{"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/arxiv-2603-01683","title":"Surgical Post-Training: Proximal On-Policy Distillation for Reasoning with Knowledge Retention","arxiv_id":"2603.01683","date":"2026-03-02","proceeding":null,"authors":["Wenye Lin","Kai Han"],"abstract":"Injecting new reasoning knowledge into Large Language Models (LLMs) via post-training often induces catastrophic forgetting. Recent studies emphasize the importance of on-policy data but suggest that KL-divergence fails to mitigate forgetting. In contrast, we show, both analytically and empirically, that the KL-constrained reward formulation actually plays a critical role in retaining knowledge during post-training. This motivates our Surgical Post-Training (SPOT), a proximal on-policy distillation framework designed to optimize reasoning efficiently while preserving prior knowledge. SPOT consists of (1) a data rectification pipeline employing an Oracle to surgically correct erroneous steps via minimal edits, generating proximal on-policy data; and (2) a reward-based binary cross-entropy objective essential for enhancing reasoning and mitigating forgetting. Empirically, with only 4k rectified math pairs, SPOT improves Qwen3-8B's accuracy by 6.2% on average across in-domain and out-of-domain tasks, requiring merely 16-minute model training on 8x H800 GPUs. Moreover, SPOT provides a superior initialization for subsequent reinforcement learning, significantly elevating the performance ceiling. Code: https://github.com/Visual-AI/SPoT","url_abs":"https://arxiv.org/abs/2603.01683","url_pdf":"https://arxiv.org/pdf/2603.01683","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2603.01683","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2603.01683"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/Visual-AI/SPoT","reach":null}],"summary":{"unverified":5},"by_repo_kind":{"found_in_text":{"samples":5,"ran":0,"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":5,"samples":[{"code_sha256_prefix":"4f0de3cab27c1825","entry":"are_lists_equal","repo":"Visual-AI/SPoT","repo_kind":"found_in_text","path":"eval/eval_connect4.py","file_url":"https://github.com/Visual-AI/SPoT/blob/HEAD/eval/eval_connect4.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":"4f0de3cab27c1825"}},{"code_sha256_prefix":"408e1f7f554c1736","entry":"eval_connect4","repo":"Visual-AI/SPoT","repo_kind":"found_in_text","path":"eval/eval_connect4.py","file_url":"https://github.com/Visual-AI/SPoT/blob/HEAD/eval/eval_connect4.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":"408e1f7f554c1736"}},{"code_sha256_prefix":"9b4f7be15a931a54","entry":"extract_coordinates","repo":"Visual-AI/SPoT","repo_kind":"found_in_text","path":"eval/eval_connect4.py","file_url":"https://github.com/Visual-AI/SPoT/blob/HEAD/eval/eval_connect4.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":"9b4f7be15a931a54"}},{"code_sha256_prefix":"f57fcdeee358817b","entry":"format_data_for_dpo","repo":"Visual-AI/SPoT","repo_kind":"found_in_text","path":"train/dpo_sft_qwen3_accelerate.py","file_url":"https://github.com/Visual-AI/SPoT/blob/HEAD/train/dpo_sft_qwen3_accelerate.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":"f57fcdeee358817b"}},{"code_sha256_prefix":"5c1a31c40a0d02d3","entry":"parse_loss_config","repo":"Visual-AI/SPoT","repo_kind":"found_in_text","path":"train/dpo_sft_qwen3_accelerate.py","file_url":"https://github.com/Visual-AI/SPoT/blob/HEAD/train/dpo_sft_qwen3_accelerate.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":"5c1a31c40a0d02d3"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}