{"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/knowledge-driven-cot-exploring-faithful","title":"Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answering","arxiv_id":"2308.13259","date":"2023-08-25","proceeding":null,"authors":["Keheng Wang","Feiyu Duan","Sirui Wang","Peiguang Li","Yunsen Xian","Chuantao Yin","Wenge Rong","Zhang Xiong"],"abstract":"Equipped with Chain-of-Thought (CoT), Large language models (LLMs) have shown impressive reasoning ability in various downstream tasks. Even so, suffering from hallucinations and the inability to access external knowledge, LLMs often come with incorrect or unfaithful intermediate reasoning steps, especially in the context of answering knowledge-intensive tasks such as KBQA. To alleviate this issue, we propose a framework called Knowledge-Driven Chain-of-Thought (KD-CoT) to verify and modify reasoning traces in CoT via interaction with external knowledge, and thus overcome the hallucinations and error propagation. Concretely, we formulate the CoT rationale process of LLMs into a structured multi-round QA format. In each round, LLMs interact with a QA system that retrieves external knowledge and produce faithful reasoning traces based on retrieved precise answers. The structured CoT reasoning of LLMs is facilitated by our developed KBQA CoT collection, which serves as in-context learning demonstrations and can also be utilized as feedback augmentation to train a robust retriever. Extensive experiments on WebQSP and ComplexWebQuestion datasets demonstrate the effectiveness of proposed KD-CoT in task-solving reasoning generation, which outperforms the vanilla CoT ICL with an absolute success rate of 8.0% and 5.1%. Furthermore, our proposed feedback-augmented retriever outperforms the state-of-the-art baselines for retrieving knowledge, achieving significant improvement in Hit and recall performance. Our code and data are released on https://github.com/AdelWang/KD-CoT/tree/main.","url_abs":"https://arxiv.org/abs/2308.13259v2","url_pdf":"https://arxiv.org/pdf/2308.13259v2.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":"knowledge-driven-cot-exploring-faithful","repo_url":"https://github.com/adelwang/kd-cot","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2308.13259","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.13259"}},"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/adelwang/kd-cot","reach":{"status":"ok"}}],"summary":{"ran":5,"ran_violates":1,"unverified":3},"by_repo_kind":{"official":{"samples":9,"ran":6,"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":9,"samples":[{"code_sha256_prefix":"7abfd90274b6deb7","entry":"MultiChoicePrompting","repo":"adelwang/kd-cot","repo_kind":"official","path":"verifier_and_cot_finetuning/data_utils/data_utils.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/verifier_and_cot_finetuning/data_utils/data_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":"7abfd90274b6deb7"}},{"code_sha256_prefix":"cba2221f35208240","entry":"cleantxt","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/GPT_inference_steps.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/GPT_inference_steps.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":"cba2221f35208240"}},{"code_sha256_prefix":"4bf8c7490aa8c35b","entry":"get_parameter_number","repo":"adelwang/kd-cot","repo_kind":"official","path":"verifier_and_cot_finetuning/data_utils/data_utils.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/verifier_and_cot_finetuning/data_utils/data_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":"4bf8c7490aa8c35b"}},{"code_sha256_prefix":"72978c28b2bf2c82","entry":"has_answer","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/Retriever/Retriever.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/Retriever/Retriever.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":"72978c28b2bf2c82"}},{"code_sha256_prefix":"7e718e1611b832e3","entry":"read_data","repo":"adelwang/kd-cot","repo_kind":"official","path":"verifier_and_cot_finetuning/data_utils/data_utils.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/verifier_and_cot_finetuning/data_utils/data_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":"7e718e1611b832e3"}},{"code_sha256_prefix":"442d4287320d335d","entry":"regex_match","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/Retriever/Retriever.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/Retriever/Retriever.py","link_basis":"plan_row","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"442d4287320d335d"}},{"code_sha256_prefix":"dcb9f54a07dc5478","entry":"process_justify_results","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/GPT_justify.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/GPT_justify.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":"dcb9f54a07dc5478"}},{"code_sha256_prefix":"6dbba84101edbf9d","entry":"process_match_results","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/GPT_justify.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/GPT_justify.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":"6dbba84101edbf9d"}},{"code_sha256_prefix":"b8284d3b260957cc","entry":"search_all","repo":"adelwang/kd-cot","repo_kind":"official","path":"pipeline/Retriever/BM25_search.py","file_url":"https://github.com/adelwang/kd-cot/blob/HEAD/pipeline/Retriever/BM25_search.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":"b8284d3b260957cc"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}