{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/in-context-sharpness-as-alerts-an-inner","title":"In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigation","arxiv_id":"2403.01548","date":"2024-03-03","proceeding":null,"authors":["Shiqi Chen","Miao Xiong","Junteng Liu","Zhengxuan Wu","Teng Xiao","Siyang Gao","Junxian He"],"abstract":"Large language models (LLMs) frequently hallucinate and produce factual errors, yet our understanding of why they make these errors remains limited. In this study, we delve into the underlying mechanisms of LLM hallucinations from the perspective of inner representations, and discover a salient pattern associated with hallucinations: correct generations tend to have sharper context activations in the hidden states of the in-context tokens, compared to the incorrect ones. Leveraging this insight, we propose an entropy-based metric to quantify the ``sharpness'' among the in-context hidden states and incorporate it into the decoding process to formulate a constrained decoding approach. Experiments on various knowledge-seeking and hallucination benchmarks demonstrate our approach's consistent effectiveness, for example, achieving up to an 8.6 point improvement on TruthfulQA. We believe this study can improve our understanding of hallucinations and serve as a practical solution for hallucination mitigation.","url_abs":"https://arxiv.org/abs/2403.01548v3","url_pdf":"https://arxiv.org/pdf/2403.01548v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"in-context-sharpness-as-alerts-an-inner","repo_url":"https://github.com/hkust-nlp/activation_decoding","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":{"status":"ok"}}],"tasks":[{"task_slug":"hallucination","task_name":"Hallucination"},{"task_slug":"truthfulqa","task_name":"TruthfulQA"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.01548","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.01548"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hkust-nlp/activation_decoding","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":2,"unverified":2},"by_repo_kind":{"official":{"samples":3,"ran":2,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":4,"samples":[{"code_sha256_prefix":"11575f75b626d3cb","entry":"extract_answer_from_output","repo":"hkust-nlp/activation_decoding","repo_kind":"official","path":"eval_knowledge_qa.py","file_url":"https://github.com/hkust-nlp/activation_decoding/blob/HEAD/eval_knowledge_qa.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"11575f75b626d3cb"}},{"code_sha256_prefix":"297c84ffca2864db","entry":"load_csv","repo":"hkust-nlp/activation_decoding","repo_kind":"official","path":"eval_tqa.py","file_url":"https://github.com/hkust-nlp/activation_decoding/blob/HEAD/eval_tqa.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"297c84ffca2864db"}},{"code_sha256_prefix":"0f14b819bc361c25","entry":"download_url","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"0f14b819bc361c25"}},{"code_sha256_prefix":"529b0ce0a10b8ff6","entry":"get_default_conv_template","repo":"hkust-nlp/activation_decoding","repo_kind":"official","path":"conversation.py","file_url":"https://github.com/hkust-nlp/activation_decoding/blob/HEAD/conversation.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"529b0ce0a10b8ff6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}