{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:MAPEPRCGA3ZBCNAML7JTZS6X4Q","short_pith_number":"pith:MAPEPRCG","schema_version":"1.0","canonical_sha256":"601e47c44606f211340c5fd33ccbd7e439bf7a5e3855c3eeb6edc12ffe26d001","source":{"kind":"arxiv","id":"2209.09513","version":2},"attestation_state":"computed","paper":{"title":"Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG","cs.MM"],"primary_cat":"cs.CL","authors_text":"Ashwin Kalyan, Kai-Wei Chang, Liang Qiu, Oyvind Tafjord, Pan Lu, Peter Clark, Song-Chun Zhu, Swaroop Mishra, Tony Xia","submitted_at":"2022-09-20T07:04:24Z","abstract_excerpt":"When answering a question, humans utilize the information available across different modalities to synthesize a consistent and complete chain of thought (CoT). This process is normally a black box in the case of deep learning models like large-scale language models. Recently, science question benchmarks have been used to diagnose the multi-hop reasoning ability and interpretability of an AI system. However, existing datasets fail to provide annotations for the answers, or are restricted to the textual-only modality, small scales, and limited domain diversity. To this end, we present Science Qu"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2209.09513","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-09-20T07:04:24Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG","cs.MM"],"title_canon_sha256":"04adf9d8f38d4cc4e14b306ee07cb7a33fc65166040aec78a2ce5c1aa59843e6","abstract_canon_sha256":"8d0d79b9ea5417e62a635b9cd3eba193c2ee92bf85764a10fb901dc512b5fdfb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:07:03.346130Z","signature_b64":"11YQLN7fjcFF4WIEvFUNuaEVRAcsqja13uCxrkO5F4oEuEkSnq+MJrhYvVstEYMLAJblDCgdKG11jrgIynutCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"601e47c44606f211340c5fd33ccbd7e439bf7a5e3855c3eeb6edc12ffe26d001","last_reissued_at":"2026-07-05T05:07:03.345713Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:07:03.345713Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG","cs.MM"],"primary_cat":"cs.CL","authors_text":"Ashwin Kalyan, Kai-Wei Chang, Liang Qiu, Oyvind Tafjord, Pan Lu, Peter Clark, Song-Chun Zhu, Swaroop Mishra, Tony Xia","submitted_at":"2022-09-20T07:04:24Z","abstract_excerpt":"When answering a question, humans utilize the information available across different modalities to synthesize a consistent and complete chain of thought (CoT). This process is normally a black box in the case of deep learning models like large-scale language models. Recently, science question benchmarks have been used to diagnose the multi-hop reasoning ability and interpretability of an AI system. However, existing datasets fail to provide annotations for the answers, or are restricted to the textual-only modality, small scales, and limited domain diversity. To this end, we present Science Qu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.09513","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2209.09513/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2209.09513","created_at":"2026-07-05T05:07:03.345774+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.09513v2","created_at":"2026-07-05T05:07:03.345774+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.09513","created_at":"2026-07-05T05:07:03.345774+00:00"},{"alias_kind":"pith_short_12","alias_value":"MAPEPRCGA3ZB","created_at":"2026-07-05T05:07:03.345774+00:00"},{"alias_kind":"pith_short_16","alias_value":"MAPEPRCGA3ZBCNAM","created_at":"2026-07-05T05:07:03.345774+00:00"},{"alias_kind":"pith_short_8","alias_value":"MAPEPRCG","created_at":"2026-07-05T05:07:03.345774+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09585","citing_title":"Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08034","citing_title":"Sci-Rho: A Multilingual Visually-Grounded Symbolic Benchmark for STEM Problems","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03569","citing_title":"When Attention Collapses: Stage-Aware Visual Token Pruning from Structure to Semantics","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29425","citing_title":"Mixture of Debaters: Learn to Debate at Architectural Level in Multi-Agent Reasoning","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25384","citing_title":"GeoMathCode: Understanding Interleaved Math-Code Reasoning for Geometry Problem Solving","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28741","citing_title":"Self-Prophetic Decoding to Unlock Visual Search in LVLMs","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30512","citing_title":"PhyDrawGen: Physically Grounded Diagram Generation from Natural Language","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2308.12067","citing_title":"MM-LIMA: Less Is More for Alignment in Multi-Modal Datasets","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2406.11354","citing_title":"Preserving Knowledge in Large Language Model with Model-Agnostic Self-Decompression","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2504.09925","citing_title":"FLARE: Fully Integration of Vision-Language Representations for Deep Cross-Modal Understanding","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2508.06038","citing_title":"Fourier Compressor: Frequency-Domain Visual Token Compression for Vision-Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2509.18150","citing_title":"Improving MLLM Training Efficiency via Stage-Aware Sparsity","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16410","citing_title":"Test-Time Hinting for Black-Box Vision-Language Models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2512.12623","citing_title":"Reasoning Within the Mind: Dynamic Multimodal Interleaving in Latent Space","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2512.15567","citing_title":"Evaluating Large Language Models in Scientific Discovery","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2503.07536","citing_title":"LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2602.15037","citing_title":"CircuChain: Disentangling Competence and Compliance in LLM Circuit Analysis","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2303.11381","citing_title":"MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12358","citing_title":"Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16930","citing_title":"CoGR-MoE: Concept-Guided Expert Routing with Consistent Selection and Flexible Reasoning for Visual Question Answering","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q","json":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q.json","graph_json":"https://pith.science/api/pith-number/MAPEPRCGA3ZBCNAML7JTZS6X4Q/graph.json","events_json":"https://pith.science/api/pith-number/MAPEPRCGA3ZBCNAML7JTZS6X4Q/events.json","paper":"https://pith.science/paper/MAPEPRCG"},"agent_actions":{"view_html":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q","download_json":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q.json","view_paper":"https://pith.science/paper/MAPEPRCG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.09513&json=true","fetch_graph":"https://pith.science/api/pith-number/MAPEPRCGA3ZBCNAML7JTZS6X4Q/graph.json","fetch_events":"https://pith.science/api/pith-number/MAPEPRCGA3ZBCNAML7JTZS6X4Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q/action/storage_attestation","attest_author":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q/action/author_attestation","sign_citation":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q/action/citation_signature","submit_replication":"https://pith.science/pith/MAPEPRCGA3ZBCNAML7JTZS6X4Q/action/replication_record"}},"created_at":"2026-07-05T05:07:03.345774+00:00","updated_at":"2026-07-05T05:07:03.345774+00:00"}