{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:SBCN7S7QJMK7VQUKJLOV6ZXOJ4","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b795235b91fe6efe961c64431f2ab871a01e03250abc48ef1198fb77424b5d43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T16:47:19Z","title_canon_sha256":"7db8783bc4159567b641a053d982b9d11196df000807f69ab69f90f1a67f6c86"},"schema_version":"1.0","source":{"id":"2506.18831","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.18831","created_at":"2026-06-19T16:09:49Z"},{"alias_kind":"arxiv_version","alias_value":"2506.18831v3","created_at":"2026-06-19T16:09:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.18831","created_at":"2026-06-19T16:09:49Z"},{"alias_kind":"pith_short_12","alias_value":"SBCN7S7QJMK7","created_at":"2026-06-19T16:09:49Z"},{"alias_kind":"pith_short_16","alias_value":"SBCN7S7QJMK7VQUK","created_at":"2026-06-19T16:09:49Z"},{"alias_kind":"pith_short_8","alias_value":"SBCN7S7Q","created_at":"2026-06-19T16:09:49Z"}],"graph_snapshots":[{"event_id":"sha256:bde5e0aa06f415bec6a25f9cc01ce2a4772d4d09bcad9b22df05484fec248306","target":"graph","created_at":"2026-06-19T16:09:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental evaluation on GSM8K demonstrates that STUPID achieves a 6% improvement in accuracy while reducing token usage by 32%, outperforming static steering baselines."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The chunk-level classifier reliably detects redundant reasoning patterns in real time and that its redundancy probability can be used as a stable error signal for the PID controller without introducing new failure modes."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"STU-PID combines a chunk-level redundancy classifier with a PID controller to dynamically modulate activation steering in LLMs, yielding 6% higher accuracy and 32% fewer tokens on GSM8K compared to static baselines."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A PID controller dynamically adjusts activation steering to cut redundant chain-of-thought steps in large language models."}],"snapshot_sha256":"adf579ab6265be845723eebbf1e5633bb3c17822e9240c62dbd92d57011e0312"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.18831/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reasoning LLMs trained with long chain-of-thought often overthink: they spend tokens on redundant reflection and transitions that inflate cost without improving accuracy. Static activation steering (e.g.\\ SEAL) suppresses such content with a fixed vector, but applies the same strength regardless of how redundant the current chunk actually is. We describe PID-steering, a training-free, decoding-time method that modulates the steering strength with a PID controller driven by a lightweight chunk-level redundancy classifier. On a subset of GSM8K with DeepSeek-R1-Distill-Qwen-1.5B, the method impro","authors_text":"Aryasomayajula Ram Bharadwaj","cross_cats":[],"headline":"A PID controller dynamically adjusts activation steering to cut redundant chain-of-thought steps in large language models.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T16:47:19Z","title":"Adaptive Activation Steering for Efficient LLM Reasoning via Closed-Loop PID Control"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.18831","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-19T07:52:07.170603Z","id":"15efd949-bac2-45fa-a367-4828f769767a","model_set":{"reader":"grok-4.3"},"one_line_summary":"STU-PID combines a chunk-level redundancy classifier with a PID controller to dynamically modulate activation steering in LLMs, yielding 6% higher accuracy and 32% fewer tokens on GSM8K compared to static baselines.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A PID controller dynamically adjusts activation steering to cut redundant chain-of-thought steps in large language models.","strongest_claim":"Experimental evaluation on GSM8K demonstrates that STUPID achieves a 6% improvement in accuracy while reducing token usage by 32%, outperforming static steering baselines.","weakest_assumption":"The chunk-level classifier reliably detects redundant reasoning patterns in real time and that its redundancy probability can be used as a stable error signal for the PID controller without introducing new failure modes."}},"verdict_id":"15efd949-bac2-45fa-a367-4828f769767a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:891203ada02a6d0f34b9fcd746c9b77b1f052a6e764c60d3d2d9443d73b10f1a","target":"record","created_at":"2026-06-19T16:09:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b795235b91fe6efe961c64431f2ab871a01e03250abc48ef1198fb77424b5d43","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-23T16:47:19Z","title_canon_sha256":"7db8783bc4159567b641a053d982b9d11196df000807f69ab69f90f1a67f6c86"},"schema_version":"1.0","source":{"id":"2506.18831","kind":"arxiv","version":3}},"canonical_sha256":"9044dfcbf04b15fac28a4add5f66ee4f16d049ce5ec2a706861fc11bdd65cf9b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9044dfcbf04b15fac28a4add5f66ee4f16d049ce5ec2a706861fc11bdd65cf9b","first_computed_at":"2026-06-19T16:09:49.750009Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:09:49.750009Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MFcSbAr4iP/QAMmP9cdTlsprTnbChaPSYWhQzQml7aDYpMcM2rjmcS0E3H7XZXk/IsOVG7t8k7YoasNIhszUCA==","signature_status":"signed_v1","signed_at":"2026-06-19T16:09:49.750442Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.18831","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:891203ada02a6d0f34b9fcd746c9b77b1f052a6e764c60d3d2d9443d73b10f1a","sha256:bde5e0aa06f415bec6a25f9cc01ce2a4772d4d09bcad9b22df05484fec248306"],"state_sha256":"b1331ddc802f5e02ef3e712268d85585b34b5227f2f5f9787917aa0d32734c5c"}