{"as_of":"2026-08-15T17:36:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:77805077c2693a7ac6fcd4d2a8b46f0b3e770d2943dc5f6ca7b569b12638118b","coverage":[{"denominator":54,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":54,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T16:55:37.731854Z","state":"measured"},{"denominator":59,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":59,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":5,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":5,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-08T10:20:42.094789Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":1,"source":"arxiv_reference","source_observed_at":"2026-08-05T02:28:24.338817Z","state":"measured"}],"external_citation_measurements":[{"count":0,"observed_at":"2026-08-05T02:28:24.338817Z","source":"arxiv_reference"}],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"cited_work":{"arxiv_id":"2508.19504","doi":"10.48550/arxiv.2508.19504","metadata_source":"arxiv_reference","pith_arxiv_id":"2508.19504","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Aegis: Taxonomy and optimizations for overcoming agent-environment failures in llm agents.arXiv preprint arXiv:2508.19504","venue":"arXiv (Cornell University)","work_id":"d4e7a45d-2b1f-4612-800d-3dd4974ba922","year":null},"citing_paper":{"arxiv_id":"2601.15808","last_updated":"2026-04-29T08:40:46Z","snapshot_observed_at":"2026-08-15T13:28:25.964823Z","submitted_at":"2026-01-22T09:47:31Z","title":"Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-05-16T12:29:06.229857Z"},"links":{"cited_paper":"/paper/2508.19504","citing_paper":"/paper/2601.15808"},"observation_digest":"sha256:ccce07e2af8132f93a8133253c9f148a5af082b886bd21f229a5d7495a503193","observation_id":"5fc8b545-4daa-4caf-9018-d1ad5b35e4dd","resolution":{"observed_at":"2026-05-16T12:30:53.870904Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"cited_work":{"arxiv_id":"2508.19504","doi":"10.48550/arxiv.2508.19504","metadata_source":"arxiv_reference","pith_arxiv_id":"2508.19504","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Aegis: Taxonomy and optimizations for overcoming agent-environment failures in llm agents.arXiv preprint arXiv:2508.19504","venue":"arXiv (Cornell University)","work_id":"d4e7a45d-2b1f-4612-800d-3dd4974ba922","year":null},"citing_paper":{"arxiv_id":"2601.15808","last_updated":"2026-04-29T08:40:46Z","snapshot_observed_at":"2026-08-15T13:28:25.964823Z","submitted_at":"2026-01-22T09:47:31Z","title":"Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-16T12:29:06.229857Z"},"links":{"cited_paper":"/paper/2508.19504","citing_paper":"/paper/2601.15808"},"observation_digest":"sha256:4ef907d8770fa113c69531fa7f80a1e202b5de6e09dac4be8281be227983d5aa","observation_id":"d7d99522-aeb0-4016-b5c3-b86392a3243b","resolution":{"observed_at":"2026-05-16T12:30:53.791034Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.19504","snapshot_observed_at":"2026-07-12T06:24:52.086585Z","title":"Aegis: Taxonomy and optimizations for overcoming agent-environment failures in llm agents,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.02882","last_updated":"2026-07-03T02:28:32Z","snapshot_observed_at":"2026-08-15T13:11:33.978918Z","submitted_at":"2026-07-03T02:28:32Z","title":"Diagnosis-Driven Automatic Repair for Agentic Workflow via Symbolic Inference","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-07-12T06:24:52.086585Z"},"links":{"cited_paper":"/paper/2508.19504","citing_paper":"/paper/2607.02882"},"observation_digest":"sha256:660557018b7da6da886408019e283885f351a96fd52237a2af56870fc598ba5b","observation_id":"415f4973-8f73-49df-8cc2-755e667d9014","resolution":{"observed_at":"2026-07-12T06:24:52.086585Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.19504","snapshot_observed_at":"2026-08-06T00:39:16.340488Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.01050","last_updated":"2026-08-02T07:32:49Z","snapshot_observed_at":"2026-08-15T05:32:28.385588Z","submitted_at":"2026-08-02T07:32:49Z","title":"Don't Offer What Can't Be Done: Deterministic Executability Gating for LLM Skill Selection at Scale","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-06T00:39:16.340488Z"},"links":{"cited_paper":"/paper/2508.19504","citing_paper":"/paper/2608.01050"},"observation_digest":"sha256:eccfac0d0df6127f2d8f882a25d1135c8ef6e385840a505a690ed69d7da0597e","observation_id":"0362ab3c-6976-46e9-864c-f84a53a5dc99","resolution":{"observed_at":"2026-08-06T00:39:16.340488Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.19504","snapshot_observed_at":"2026-08-08T10:20:42.094789Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.05573","last_updated":"2026-08-06T03:52:37Z","snapshot_observed_at":"2026-08-13T00:53:53.028383Z","submitted_at":"2026-08-06T03:52:37Z","title":"SkillTV-Bench: Benchmarking How Well Judges Perform on Skill-Augmented Agentic Execution","version":1},"reference_index":76,"source":"arxiv_source","source_observed_at":"2026-08-08T10:20:42.094789Z"},"links":{"cited_paper":"/paper/2508.19504","citing_paper":"/paper/2608.05573"},"observation_digest":"sha256:347d8a2a2f5b36eef9f79ad19c8b13896269719cfacf9f11e355f137d8a28797","observation_id":"fe62e75e-1728-43a8-81e8-54901ceac076","resolution":{"observed_at":"2026-08-08T10:20:42.094789Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2508.19504/citation-record","integrity":"/paper/2508.19504/integrity","json":"/paper/2508.19504/citation-record.json","paper":"/paper/2508.19504"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.416855Z","title":"Introducing Claude 4, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.416855Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:6de4728acd3f8b26ada96f7333875544d2226f87a145db022de59441d1b401e9","observation_id":"3ef801cf-fb1f-422b-aa89-a8d164e814e4","resolution":{"observed_at":"2026-08-15T16:55:37.416855Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.665039Z","title":null,"venue":null,"work_id":"9f622861-d2c9-4dcb-9613-300aa48c32a3","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.422743Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:357f1c65f8ec49118458e0ef5c1ff1ec448bf55890c4f2208e429e602bd47b25","observation_id":"c9d2a663-2573-4c47-961a-0abca8e1b22f","resolution":{"observed_at":"2026-08-15T16:55:38.673162Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.428785Z","title":"Pan, Shuyi Yang, Lakshya A","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.428785Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:4729e59bc582822521b7cfc8eb4f1b05b3eb9f033c9dba68327631e86fc08ba1","observation_id":"7d3a6ad5-86c9-4254-bb72-d37417dd9c59","resolution":{"observed_at":"2026-08-15T16:55:37.428785Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.620821Z","title":"Standard benchmarks fail – auditing llm agents in finance must prioritize risk, 2025","venue":null,"work_id":"89ac50da-f41d-4b69-a522-a785159069a8","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.434225Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:367055f7bd23d14732c0e6f972a3c9488f6f9d071ce83e430962dd17ccb6145b","observation_id":"95738c90-7f03-43c3-8da9-e6cc869e374a","resolution":{"observed_at":"2026-08-15T16:55:38.627995Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.597848Z","title":"Understanding fine-tuning, 2024","venue":null,"work_id":"da1bb7eb-130d-4df0-828a-f4de277081ad","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.441930Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:8fa9ef541f268b0b83b14b44c868d2d153c2a9b0fe7f44b388481949aa6813f3","observation_id":"d07e8341-0898-4bfe-9412-b09f218a05ba","resolution":{"observed_at":"2026-08-15T16:55:38.605034Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.447143Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.447143Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:6139f991a5a0b3189d3736f83e0115b902af02d4be095330354fd3805784a5e7","observation_id":"b33fa9a6-3ad3-4021-b6b5-586d89b71025","resolution":{"observed_at":"2026-08-15T16:55:37.447143Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.565888Z","title":"Trail: Trace reasoning and agentic issue localization, 2025","venue":null,"work_id":"23360877-4f32-48c0-a9cb-c21c0b4cede8","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.453693Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:4ca24b9d9670d113518cefa694227225f8680978de9fef59c47f1022597215d3","observation_id":"06885efb-2913-458a-91cf-c614decc690b","resolution":{"observed_at":"2026-08-15T16:55:38.571423Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.540592Z","title":"An overview of hierarchical task network planning, 2014","venue":null,"work_id":"b23d396f-a890-4db5-8ee1-4d276de074db","year":2014},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.458822Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:d49186d86dbc22f83cbef53f9acb808e2a289a1fb6063fab25db81a0a4b118e0","observation_id":"ef140434-5956-4dab-b977-06dc2d5e1c64","resolution":{"observed_at":"2026-08-15T16:55:38.548764Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.520528Z","title":"Advanced version of Gemini with deep think officially achieves gold-medal standard at the international mathematical olympiad, 2025","venue":null,"work_id":"c6cf8422-eb9d-49e2-a443-4a3d0ef375c1","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.464395Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:7c15939069244871299c7d94a1c81ff49864ad460e8f3debf46ce9cda1363486","observation_id":"03e65dbb-39b5-4f01-a31e-f49d4ec30653","resolution":{"observed_at":"2026-08-15T16:55:38.526571Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.501355Z","title":"Gemini 2.5 Pro, 2025","venue":null,"work_id":"25f29cd6-4327-4ad8-b80e-ac8a147f4892","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.470056Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:22d2b71f9469471b06933e763a22214f36b728d28205ef4ae9ce77892039b994","observation_id":"55716734-b961-46ad-b8ee-6f8c3877c632","resolution":{"observed_at":"2026-08-15T16:55:38.508279Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.481694Z","title":"Intelligent virtual assistants with llm-based process automation, 2023","venue":null,"work_id":"c5a02069-a255-4ab9-b783-f169b032ba45","year":2023},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.474856Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:58fb95c910e08dc017c9b76b92cc4f023102374b5976812cceb0b95784544eea","observation_id":"0bb603f2-0b8b-4ba8-bf0d-c806e9e58aa3","resolution":{"observed_at":"2026-08-15T16:55:38.488549Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.456097Z","title":"Crmarena: Understanding the capacity of llm agents to perform professional crm tasks in realistic environments","venue":null,"work_id":"13a6f3f6-7c1f-4c75-9ee7-a30723a4cda2","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.480137Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:8fcc1c90e88f7ed46d258da5c48196a9f1c8a3ea10cbb09db7fc203dcb4c3d27","observation_id":"0395e57d-05cc-4ca4-ae32-87b1d78c04a2","resolution":{"observed_at":"2026-08-15T16:55:38.461334Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.438076Z","title":"Crmarena-pro: Holistic assessment of llm agents across diverse business scenarios and interactions, 2025","venue":null,"work_id":"9f3ec612-93d5-4c43-ae38-5b2f73df9777","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.485276Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:d4aa974852d917da167d0e180dbc16feb0d670185edea28dc94744e277babd29","observation_id":"7caae0a4-23c6-4fd6-81f2-a5532ca3cae5","resolution":{"observed_at":"2026-08-15T16:55:38.444768Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.490010Z","title":"Black, Gloria Geng, Danny Park, James Zou, Andrew Y","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.490010Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:2c8d764e2c3a758f5a16991241b6fe49f49180ee70cb2f56e80aafeab35072a2","observation_id":"0e59bf87-cc1c-4928-8155-4ceeed537d3e","resolution":{"observed_at":"2026-08-15T16:55:37.490010Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.408530Z","title":"Proxyllm : Llm-driven framework for customer support through text-style transfer, 2024","venue":null,"work_id":"aad75344-01e5-423b-ba8c-d28893e271b0","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.496877Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:892818dd9e59666b43ad3d985ac12a61bde25d2686d5ffc3a052159e92ee0617","observation_id":"57e38973-8a74-4f7f-8766-4195e0d5afb5","resolution":{"observed_at":"2026-08-15T16:55:38.413699Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.390395Z","title":"The hidden cost of ignoring llm failures, 2025","venue":null,"work_id":"a8676261-59d9-4978-865c-b55cd7dc7f1d","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.504464Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:d9d41c3098e23ea50e1f2ad9cdd47c5698c371376335a1f56bab4b69959eefe3","observation_id":"fdede9fb-e2b7-4f91-a776-67259523cc87","resolution":{"observed_at":"2026-08-15T16:55:38.397428Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.510468Z","title":"Scaling laws for forgetting when fine-tuning large language models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.510468Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:ce4dfaa948fe85719756d950426ef1277a4f870c8c299cfd40a4b3e4d22010d7","observation_id":"e144a0ca-a2d4-46cb-95ff-8c0d561b0632","resolution":{"observed_at":"2026-08-15T16:55:37.510468Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.361071Z","title":"Ziegler, Elizabeth Barnes, and Lawrence Chan","venue":null,"work_id":"e518be9d-a5dd-43d3-bd54-75279c510549","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.515949Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:f836a3f7b98caf44dec680c631645470717398e49fb7c2682bd136fb951ea770","observation_id":"bd6d8d84-2a3a-4a63-a016-4a8f68cbc1a4","resolution":{"observed_at":"2026-08-15T16:55:38.366426Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.529018Z","title":"Llms get lost in multi-turn conversation, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.529018Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:e7cb82ff2151e96f4b7dd78b6d07e395985f709259b9618e00a591eb7dd72204","observation_id":"ca72e421-c0f5-4afa-9259-0a9e11e6e6ca","resolution":{"observed_at":"2026-08-15T16:55:37.529018Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.332250Z","title":"Spider 2.0: Evaluating language models on real-world enterprise text-to-sql workflows, 2025","venue":null,"work_id":"fcfb9074-1a8d-4429-9eed-63da7e8b5304","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.538973Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:93cda3359e904f78d52e828eaaced113eeafb871180eb148eb5352fc2b96d7c8","observation_id":"5ecae5b0-cc99-491f-8ae7-c96c3599424c","resolution":{"observed_at":"2026-08-15T16:55:38.337488Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.315322Z","title":"Personal llm agents: Insights and survey about the capability, efficiency and security, 2024","venue":null,"work_id":"90d0caa3-d053-48b5-849e-bb543708712e","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.546731Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:c2b84de6fa64466e0f0a410f701e4769e42ff86b4596c1ba5d84851f270f9ad1","observation_id":"2f69c910-61af-4d96-b78c-ffdc1687e598","resolution":{"observed_at":"2026-08-15T16:55:38.320800Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.298240Z","title":"Toolace: Winning the points of llm function calling, 2025","venue":null,"work_id":"5a05d32a-3391-43da-80d4-ac8e3e173efc","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.551836Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:0cd41b3b889d23eb071bbe965180b3cfee96de6a52eed15c45c516c6a78e4bc4","observation_id":"7c6b11b0-1b4c-4877-a63a-6cb814cff3f4","resolution":{"observed_at":"2026-08-15T16:55:38.303684Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.557325Z","title":"Agentbench: Evaluating llms as agents, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.557325Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:c98f90aef1f6e74ef99437c34efbc2756547284ae67109f66f8cf1670c2f2b64","observation_id":"1a09c289-47f2-4602-be10-41cd2cd88d72","resolution":{"observed_at":"2026-08-15T16:55:37.557325Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.563880Z","title":"Toolsandbox: A stateful, conversational, interactive evaluation benchmark for llm tool use capabilities, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.563880Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:c4c8462c10c93020e3e3ebc34b07242b3d8116fb82dc04d488b980a939f2d906","observation_id":"bd1be5db-24fb-44c3-80fe-bdc267c5b23a","resolution":{"observed_at":"2026-08-15T16:55:37.563880Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.569625Z","title":"An empirical study of catastrophic forgetting in large language models during continual fine-tuning, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.569625Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:d86ba0e39bb67d46ee67d282da93400b50bbaac0bb67100ef8bf553ff7b8d53c","observation_id":"85d4371f-5214-44a2-8211-ff4bfb9834d2","resolution":{"observed_at":"2026-08-15T16:55:37.569625Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.246924Z","title":"Rag vs fine tuning: How to choose the right method, 2024","venue":null,"work_id":"611607eb-5740-407e-b7a8-e2e1826b991c","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.574760Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:26c1c024268f5365238cd6e1dc7f35459fbfbe55b851cd2d0c94264259f0aa30","observation_id":"b1d6eaab-dd77-47cd-958e-9f8fefa54fa8","resolution":{"observed_at":"2026-08-15T16:55:38.254754Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.230308Z","title":"Bfcl v3 multi-turn and multi-step function calling evaluation, 2024","venue":null,"work_id":"c0fd3cf1-cba3-4250-9e61-58974ac997be","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.582996Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:a8c859a3366b6696231c702b882e29b1ab95e49e1e34ad5dbfef3be3f69ff862","observation_id":"e13c5f09-4af9-4383-b5ab-49718d09c88f","resolution":{"observed_at":"2026-08-15T16:55:38.235721Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.211623Z","title":"Introducing meta llama 3: The most capable openly available llm to date, 2024","venue":null,"work_id":"1c6ceede-45b4-402e-9fda-500dca7b288e","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.588997Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:9cb23e10e60f52b7b2ac826ed7bdb4972a868a1a236bf0a1fd4a8e8b2f320ca3","observation_id":"637fe108-dc34-4375-8ba4-ffe306ee80a9","resolution":{"observed_at":"2026-08-15T16:55:38.217492Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.193289Z","title":"Agentmisalignment: Measuring the propensity for misaligned behaviour in llm-based agents, 2025","venue":null,"work_id":"148f1ac9-fcd2-4dee-a0cf-0f84ac548b71","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.594260Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:17a6714a35fbc0235cbfe18c6126e5c3c11cf62c16b42e864f0f75863d3aa968","observation_id":"6bbb5704-3174-4870-8668-480c0f9b51af","resolution":{"observed_at":"2026-08-15T16:55:38.199706Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.175191Z","title":"Hierarchical task network (htn) planning, 2012","venue":null,"work_id":"0885087f-6e50-4354-a451-810371fdb11c","year":2012},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.600610Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:b1e731f5ad198becd7ad095531fa0d59d18056594c106f5455b96154d76597dd","observation_id":"a470e2ee-26ef-4b32-81b8-749fabf246ba","resolution":{"observed_at":"2026-08-15T16:55:38.181912Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.154741Z","title":"Learning to reason with LLMs, 2024","venue":null,"work_id":"e55e7b84-d6bf-4a3e-b8e7-47f65ba4d773","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.605894Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:ab9a8fcdc6105fb8c6fd814ddbf3448893abb02f76ecf87df60a737f9b57e4d8","observation_id":"90ffb51c-b100-4ebf-b81b-bc088dde57eb","resolution":{"observed_at":"2026-08-15T16:55:38.160480Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.137490Z","title":"Introducing ChatGPT agent: bridging research and action, 2025","venue":null,"work_id":"c5b06c75-d310-4abc-a128-14faa033dc75","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.610850Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:889f51bc019f55550d7017418d5f02837c1b50e8c0e2b5dce51ac353054fb811","observation_id":"cc8a9a47-72d5-4cfb-9372-2d2868464160","resolution":{"observed_at":"2026-08-15T16:55:38.142995Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.615750Z","title":"Introducing GPT-4.1 in the API, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.615750Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:2024b16317d80dbb315515783a914b7ac437a48a45408f1eb2ebcc45249285ac","observation_id":"e5ba2b97-bb22-434a-b030-91154076c2ea","resolution":{"observed_at":"2026-08-15T16:55:37.615750Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.623163Z","title":"Introducing openai o3 and o4-mini, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.623163Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:183f6d17704075802397a2c4d06c1f06cf5d58475e6e01693951c8fc3c6b6d08","observation_id":"c3814712-724d-4a9a-abd2-0547824b58e8","resolution":{"observed_at":"2026-08-15T16:55:37.623163Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.097690Z","title":"Gonzalez","venue":null,"work_id":"8e2ea929-4102-4093-9b47-7a9d270dd672","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.629516Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:305bde6686f5b2f55ee6dbdcf9b7cd694fec7f1fbca8c469fe4230f10ee7d5b1","observation_id":"752bf8fb-010e-4ed6-b9f0-984315c1fb9c","resolution":{"observed_at":"2026-08-15T16:55:38.102687Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.080197Z","title":"Apigen-mt: Agentic pipeline for multi- turn data generation via simulated agent-human interplay, 2025","venue":null,"work_id":"4a6e776f-7259-4ef8-8c6d-28eee7f13589","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.634908Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:bbb699cb8157539fdea062feee4e9c38d4ccb710c028b3a131900986e2aea18e","observation_id":"405427b8-a266-4296-9628-8cbde688e969","resolution":{"observed_at":"2026-08-15T16:55:38.085846Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.641141Z","title":"Maddison, and Tatsunori Hashimoto","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.641141Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:42380294f0e0bcfe35bcc8e695a58ef9ba3d176bc94b0651e2fb7c2e414ab4df","observation_id":"8ea7308f-1ccf-4ec7-abca-88210b81f085","resolution":{"observed_at":"2026-08-15T16:55:37.641141Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.646474Z","title":"Large language models can be easily distracted by irrelevant context, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.646474Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:be8babd4cd9a37044e8cd1134ee56bde69c4c6b2454a271097466617023dea4a","observation_id":"8252b4bf-71d9-42df-ab48-2c08061ad7be","resolution":{"observed_at":"2026-08-15T16:55:37.646474Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.041950Z","title":"Direct multi-turn preference optimization for language agents, 2025","venue":null,"work_id":"58e7f00f-08ce-4b8d-9042-1bec11af589d","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.651109Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:b65bb4b288b640a0f42b29dad5e3db972a75fb6fb40d9059a174f47420c88e3d","observation_id":"bd18b089-092a-4ce2-9752-109faa4b6759","resolution":{"observed_at":"2026-08-15T16:55:38.047206Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.025282Z","title":"From commands to prompts: Llm-based semantic file system for aios, 2025","venue":null,"work_id":"fc0afbee-4a3c-41d0-b5e7-beb48ac6fdaf","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.656065Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:a0232b204df3ea3c632687746af6c9368d1cc59ce9ea322e25b0dac268816cd0","observation_id":"d05cd24f-6f7d-44fa-b2fe-5366a028ef4d","resolution":{"observed_at":"2026-08-15T16:55:38.030778Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:38.001610Z","title":"Salesforce study finds llm agents fail 65","venue":null,"work_id":"26861b2d-6da3-4fea-bb23-bfc3291e7a94","year":null},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.660688Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:e9027da1dd96ec402ac9bc52d9e8e8a890d56a4e18bd56b2dcff737a4d8a471f","observation_id":"ced4d35f-fc13-4ea1-b5f1-0492aca96682","resolution":{"observed_at":"2026-08-15T16:55:38.008380Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.985633Z","title":"Understanding fine-tuning, 2025","venue":null,"work_id":"54a553fb-508d-455c-81e1-060aee547552","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.665811Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:60ff6909dee37e60867a26f0f64ffdc1b9dc8825897879cc948f81d2c18346a9","observation_id":"4e4e5159-91da-4e2b-a9b3-c9a40fc2af05","resolution":{"observed_at":"2026-08-15T16:55:37.990438Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.671037Z","title":"Understanding the weakness of large language model agents within a complex android environment, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.671037Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:05a5932f9cd02337f7f4bfbbb54c9bcbe47b4be159d2cccb69afec8d72577d0d","observation_id":"fc5c1f65-5e1e-4e47-a556-fd8833c077f4","resolution":{"observed_at":"2026-08-15T16:55:37.671037Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.677185Z","title":"Jimenez, Alexander Wettig, Kilian Lieret, Shunyu Yao, Karthik Narasimhan, and Ofir Press","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.677185Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:b8793b4f312f59486324698c7a0cbbfd09aa95f0735bac0c9bad8b54a202c41e","observation_id":"3d521e6e-ec4f-4ad4-8e44-3421da47aba8","resolution":{"observed_at":"2026-08-15T16:55:37.677185Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.944036Z","title":"Agentoccam: A simple yet strong baseline for llm-based web agents, 2025","venue":null,"work_id":"f9e59bf7-0a71-4f81-8cbe-6d055f35e00f","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.685606Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:a138128ab88d4c7a80a3cc2939a48e9e31031fef37299be71cc55e490fb96ea7","observation_id":"190a0f47-0427-4534-a7e9-b9f313b5da66","resolution":{"observed_at":"2026-08-15T16:55:37.950665Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.925290Z","title":"𝜏-bench: A benchmark for tool-agent-user interaction in real-world domains, 2024","venue":null,"work_id":"7f74d702-30b7-45b2-a2e8-77d98fa102f3","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.690559Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:5ad6556c5a9e5b3efcd538a06f51c5919a223fde75f5b5704c73dbb466eeb2dc","observation_id":"d3a11972-18d3-439e-a79e-e70d1b9654de","resolution":{"observed_at":"2026-08-15T16:55:37.930898Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.695229Z","title":"Spider: A large-scale human-labeled dataset for complex and cross-domain semantic parsing and text-to-sql task, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.695229Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:e0a78e1c441b6d2295f03a96a3175c4107f8d05dc276d6e7b677582fab662357","observation_id":"93d28927-8b46-4f7f-8970-dc9d70d04cb1","resolution":{"observed_at":"2026-08-15T16:55:37.695229Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.891564Z","title":"Actionstudio: A lightweight framework for data and training of large action models, 2025","venue":null,"work_id":"ca682e60-59b6-4474-8960-8ab50f1ecb01","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.702222Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:64d3e0d015c46c3db02d252b4698d78d76c457a90ac87487807997171246360b","observation_id":"a4ea67a2-5074-444c-a2dd-d2209624c79e","resolution":{"observed_at":"2026-08-15T16:55:37.897884Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.872092Z","title":"xlam: A family of large action models to empower ai agent systems, 2024","venue":null,"work_id":"8a584880-2711-4c4a-80df-d0b66555c61d","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.707597Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:1b776071e39ed0a7be9f1a5e4217abbff8c73d072171a2338fce2eac35b2c6c5","observation_id":"b1462bc0-7ef7-4250-8806-9edf7565e55f","resolution":{"observed_at":"2026-08-15T16:55:37.878450Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.847612Z","title":"Which agent causes task failures and when? on automated failure attribution of llm multi-agent systems, 2025","venue":null,"work_id":"68dbd3ce-3d33-4697-8ca8-02f707b01af7","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.712287Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:ec86ae45415397d389fe895f8820e35604553787a0fa0d296a09e8e1125938fc","observation_id":"84cee6cb-77ee-4680-a0d3-a11da064c785","resolution":{"observed_at":"2026-08-15T16:55:37.856997Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.823649Z","title":"General modular harness for llm agents in multi-turn gaming envi- ronments, 2025","venue":null,"work_id":"451d9709-144c-4634-bf18-595f896919ae","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.717031Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:d10df046651e99cb7994dda9c651215c37b42bdc2f60addb880b983b76f29914","observation_id":"ae5a4be7-ffa5-41be-a566-094a4b63f5fb","resolution":{"observed_at":"2026-08-15T16:55:37.829622Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.803520Z","title":"Agent-safetybench: Evaluating the safety of llm agents, 2025","venue":null,"work_id":"43bb4e9c-697d-437a-85bc-17c8cd74ca83","year":2025},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.721941Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:f9078faadf5f816f710ed92cb688ee05c4fedc5382f2761df10515910e1fd3e2","observation_id":"c5bb2f69-9050-4eee-8180-6390711571c8","resolution":{"observed_at":"2026-08-15T16:55:37.809641Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.727098Z","title":"Xu, Hao Zhu, Xuhui Zhou, Robert Lo, Abishek Sridhar, Xianyi Cheng, Tianyue Ou, Yonatan Bisk, Daniel Fried, Uri Alon, and Graham Neubig","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.727098Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:4fb39808378691e5e98c5d8787764be1a47306c1681c0b08434a251f7eac50ac","observation_id":"dd041582-31a9-4b24-8f56-458fb152ceb9","resolution":{"observed_at":"2026-08-15T16:55:37.727098Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T16:55:37.770581Z","title":"Large language model enhanced text-to-sql generation: A survey, 2024","venue":null,"work_id":"a3e6bd44-b223-4c44-9ae5-c9ccc0974a38","year":2024},"citing_paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-15T16:55:37.731854Z"},"links":{"citing_paper":"/paper/2508.19504"},"observation_digest":"sha256:e9380d1cb853026ec0afd4dcef99db62b1e3cd021f56d85db365bb35dfe9bb88","observation_id":"ff30ec0a-cfab-4912-baf9-bc08e2eeea00","resolution":{"observed_at":"2026-08-15T16:55:37.778186Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-15T06:32:42.880941+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2508.19504","last_updated":"2025-08-27T01:29:46Z","latest_version":1,"primary_category":"cs.MA","snapshot_observed_at":"2026-08-15T16:49:36.013627Z","submitted_at":"2025-08-27T01:29:46Z","title":"Aegis: Taxonomy and Optimizations for Overcoming Agent-Environment Failures in LLM Agents"},"reference_resolution":{"displayed":54,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":18,"verified_exact":0,"verified_fuzzy":36},"total_outbound_references":54},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 15 August 2026, this Paper Citation Record lists 54 of 54 outbound references and 5 inbound Pith citation observations for arXiv:2508.19504."}