{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:V4HOK5QSDA64FGYYEOSZIPJGNU","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":"f618e92703cc0ab7c2559f1935c69f19ef4db830d08e82e0a96a89d3c64c554e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T05:33:56Z","title_canon_sha256":"e48e8330ec0263c9ded7309878065d18bc40cf4e239f75120e510cdbe95fe3c8"},"schema_version":"1.0","source":{"id":"2506.09408","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09408","created_at":"2026-07-05T11:19:36Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09408v1","created_at":"2026-07-05T11:19:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09408","created_at":"2026-07-05T11:19:36Z"},{"alias_kind":"pith_short_12","alias_value":"V4HOK5QSDA64","created_at":"2026-07-05T11:19:36Z"},{"alias_kind":"pith_short_16","alias_value":"V4HOK5QSDA64FGYY","created_at":"2026-07-05T11:19:36Z"},{"alias_kind":"pith_short_8","alias_value":"V4HOK5QS","created_at":"2026-07-05T11:19:36Z"}],"graph_snapshots":[{"event_id":"sha256:56917b60136af7cae01d55b4b403ac517fe95e25270cf513fc8a05740623beb0","target":"graph","created_at":"2026-07-05T11:19:36Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.09408/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive performance on multiple-choice question answering (MCQA) benchmarks, yet they remain highly vulnerable to minor input perturbations. In this paper, we introduce and evaluate Token Constraint Decoding (TCD). This simple yet effective inference-time algorithm enforces alignment between token-level predictions to enhance robustness in noisy settings. Through extensive experiments on CommonsenseQA, MMLU, and MMLU-Pro, we show that TCD, especially when paired with prompt engineering (PE) fixes, significantly restores performance degraded by ","authors_text":"Bing-Cheng Xie, Bing-Jia Tan, Hao-Yuan Chen, Jui-Ming Yao, Sheng-Wei Peng, Shun-Feng Su, Zi-Xian Tang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T05:33:56Z","title":"Token Constraint Decoding Improves Robustness on Question Answering for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09408","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:71b8c57dc7f304023ef79742497e80855d5d6760b6216db5b83ea4c1026dcf01","target":"record","created_at":"2026-07-05T11:19:36Z","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":"f618e92703cc0ab7c2559f1935c69f19ef4db830d08e82e0a96a89d3c64c554e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T05:33:56Z","title_canon_sha256":"e48e8330ec0263c9ded7309878065d18bc40cf4e239f75120e510cdbe95fe3c8"},"schema_version":"1.0","source":{"id":"2506.09408","kind":"arxiv","version":1}},"canonical_sha256":"af0ee57612183dc29b1823a5943d266d07c04b0614e3a83373cd52adaaca2f6c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"af0ee57612183dc29b1823a5943d266d07c04b0614e3a83373cd52adaaca2f6c","first_computed_at":"2026-07-05T11:19:36.580842Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:19:36.580842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L5CWxdqLfidhkhvLBV8iBhvzMVGZ1V7CyDDJq3c48Ne7UAJp0jXDeWMea5DosKDTxBDMfu372nF+npJA5hR0Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:19:36.581282Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.09408","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:71b8c57dc7f304023ef79742497e80855d5d6760b6216db5b83ea4c1026dcf01","sha256:56917b60136af7cae01d55b4b403ac517fe95e25270cf513fc8a05740623beb0"],"state_sha256":"226a6987a970ada48b4893d20d4df50e939983ad04f805a2e767979c832f2a3a"}