{"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/how-interpretable-are-reasoning-explanations","title":"How Interpretable are Reasoning Explanations from Prompting Large Language Models?","arxiv_id":"2402.11863","date":"2024-02-19","proceeding":null,"authors":["Wei Jie Yeo","Ranjan Satapathy","Rick Siow Mong Goh","Erik Cambria"],"abstract":"Prompt Engineering has garnered significant attention for enhancing the performance of large language models across a multitude of tasks. Techniques such as the Chain-of-Thought not only bolster task performance but also delineate a clear trajectory of reasoning steps, offering a tangible form of explanation for the audience. Prior works on interpretability assess the reasoning chains yielded by Chain-of-Thought solely along a singular axis, namely faithfulness. We present a comprehensive and multifaceted evaluation of interpretability, examining not only faithfulness but also robustness and utility across multiple commonsense reasoning benchmarks. Likewise, our investigation is not confined to a single prompting technique; it expansively covers a multitude of prevalent prompting techniques employed in large language models, thereby ensuring a wide-ranging and exhaustive evaluation. In addition, we introduce a simple interpretability alignment technique, termed Self-Entailment-Alignment Chain-of-thought, that yields more than 70\\% improvements across multiple dimensions of interpretability. Code is available at https://github.com/SenticNet/CoT_interpretability","url_abs":"https://arxiv.org/abs/2402.11863v3","url_pdf":"https://arxiv.org/pdf/2402.11863v3.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":"how-interpretable-are-reasoning-explanations","repo_url":"https://github.com/wj210/cot_interpretability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"prompt-engineering","task_name":"Prompt Engineering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2402.11863","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.11863"}},"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/wj210/cot_interpretability","reach":{"status":"ok"}}],"summary":{"ran":13,"unverified":3},"by_repo_kind":{"official":{"samples":16,"ran":13,"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":16,"samples":[{"code_sha256_prefix":"46a8fc7b8d455553","entry":"create_entailment_prompt","repo":"wj210/cot_interpretability","repo_kind":"official","path":"template/prompt_template.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/template/prompt_template.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"46a8fc7b8d455553"}},{"code_sha256_prefix":"c010705f29317501","entry":"cross_check_ans","repo":"wj210/cot_interpretability","repo_kind":"official","path":"openai_gen.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/openai_gen.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c010705f29317501"}},{"code_sha256_prefix":"882636a19c02bd1c","entry":"format_explanation","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/data_collator_pinto.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/data_collator_pinto.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"882636a19c02bd1c"}},{"code_sha256_prefix":"9bac83066a04172a","entry":"format_input","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/data_collator_pinto.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/data_collator_pinto.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9bac83066a04172a"}},{"code_sha256_prefix":"d9e7d08864a2514a","entry":"format_llama_prompt","repo":"wj210/cot_interpretability","repo_kind":"official","path":"template/prompt_template.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/template/prompt_template.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d9e7d08864a2514a"}},{"code_sha256_prefix":"9f162717bd58428a","entry":"get_label","repo":"wj210/cot_interpretability","repo_kind":"official","path":"openai_gen.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/openai_gen.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9f162717bd58428a"}},{"code_sha256_prefix":"d9efb3b6e8fe970a","entry":"get_label_tensor","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/data_collator_pinto.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/data_collator_pinto.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d9efb3b6e8fe970a"}},{"code_sha256_prefix":"91902d78bfbae499","entry":"get_logger","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"91902d78bfbae499"}},{"code_sha256_prefix":"4e7d204fa2924e1e","entry":"preprocess_all","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4e7d204fa2924e1e"}},{"code_sha256_prefix":"5ff002e9bf012277","entry":"preprocess_input","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/utils.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5ff002e9bf012277"}},{"code_sha256_prefix":"9e0cd81e8cf96360","entry":"process_answer","repo":"wj210/cot_interpretability","repo_kind":"official","path":"template/refine_template.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/template/refine_template.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9e0cd81e8cf96360"}},{"code_sha256_prefix":"6b43934be1427d31","entry":"process_feedback","repo":"wj210/cot_interpretability","repo_kind":"official","path":"template/refine_template.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/template/refine_template.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6b43934be1427d31"}},{"code_sha256_prefix":"23df1880ebfb46ca","entry":"prompt_input","repo":"wj210/cot_interpretability","repo_kind":"official","path":"template/prompt_template.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/template/prompt_template.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"23df1880ebfb46ca"}},{"code_sha256_prefix":"f86199c589be7bff","entry":"clear_pad","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/model_utils.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/model_utils.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":"f86199c589be7bff"}},{"code_sha256_prefix":"78211d8377e50f1d","entry":"faithfulness_test","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/model_utils.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/model_utils.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":"78211d8377e50f1d"}},{"code_sha256_prefix":"b2d1503e1d9306ee","entry":"load_raw_dataset","repo":"wj210/cot_interpretability","repo_kind":"official","path":"utils/data_helper.py","file_url":"https://github.com/wj210/cot_interpretability/blob/HEAD/utils/data_helper.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":"b2d1503e1d9306ee"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}