{"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/transformer-copilot-learning-from-the-mistake","title":"Transformer Copilot: Learning from The Mistake Log in LLM Fine-tuning","arxiv_id":"2505.16270","date":"2025-05-22","proceeding":null,"authors":["Jiaru Zou","Yikun Ban","Zihao Li","Yunzhe Qi","Ruizhong Qiu","Ling Yang","Jingrui He"],"abstract":"Large language models are typically adapted to downstream tasks through supervised fine-tuning on domain-specific data. While standard fine-tuning focuses on minimizing generation loss to optimize model parameters, we take a deeper step by retaining and leveraging the model's own learning signals, analogous to how human learners reflect on past mistakes to improve future performance. We first introduce the concept of Mistake Log to systematically track the model's learning behavior and recurring errors throughout fine-tuning. Treating the original transformer-based model as the Pilot, we correspondingly design a Copilot model to refine the Pilot's inference performance via logits rectification. We name the overall Pilot-Copilot framework the Transformer Copilot, which introduces (i) a novel Copilot model design, (ii) a joint training paradigm where the Copilot continuously learns from the evolving Mistake Log alongside the Pilot, and (iii) a fused inference paradigm where the Copilot rectifies the Pilot's logits for enhanced generation. We provide both theoretical and empirical analyses on our new learning framework. Experiments on 12 benchmarks spanning commonsense, arithmetic, and recommendation tasks demonstrate that Transformer Copilot consistently improves performance by up to 34.5%, while introducing marginal computational overhead to Pilot models and exhibiting strong scalability and transferability.","url_abs":"https://arxiv.org/abs/2505.16270v1","url_pdf":"https://arxiv.org/pdf/2505.16270v1.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":"transformer-copilot-learning-from-the-mistake","repo_url":"https://github.com/jiaruzouu/transformercopilot","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2505.16270","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2505.16270"}},"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/jiaruzouu/transformercopilot","reach":null}],"summary":{"ran_honours":3,"ran_draft_wrong":2},"by_repo_kind":{"official":{"samples":4,"ran":4,"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":1,"samples":[{"code_sha256_prefix":"a1b46b3f84e76f6a","entry":"apply_rotary_pos_emb","repo":"jiaruzouu/transformercopilot","repo_kind":"official","path":"src/Decoder_Only/copilot/modeling_flax_llama.py","file_url":"https://github.com/jiaruzouu/transformercopilot/blob/HEAD/src/Decoder_Only/copilot/modeling_flax_llama.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a1b46b3f84e76f6a"}},{"code_sha256_prefix":"660ecba66bbbf2c0","entry":"create_sinusoidal_positions","repo":"jiaruzouu/transformercopilot","repo_kind":"official","path":"src/Decoder_Only/copilot/modeling_flax_llama.py","file_url":"https://github.com/jiaruzouu/transformercopilot/blob/HEAD/src/Decoder_Only/copilot/modeling_flax_llama.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"660ecba66bbbf2c0"}},{"code_sha256_prefix":"56e674752db00649","entry":"generate_prompt","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"56e674752db00649"}},{"code_sha256_prefix":"32ae10f6fdd7e688","entry":"load_data","repo":"jiaruzouu/transformercopilot","repo_kind":"official","path":"src/Decoder_Only/finetune.py","file_url":"https://github.com/jiaruzouu/transformercopilot/blob/HEAD/src/Decoder_Only/finetune.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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"32ae10f6fdd7e688"}},{"code_sha256_prefix":"6b3f1075d5cc01c4","entry":"rotate_half","repo":"jiaruzouu/transformercopilot","repo_kind":"official","path":"src/Decoder_Only/copilot/modeling_flax_llama.py","file_url":"https://github.com/jiaruzouu/transformercopilot/blob/HEAD/src/Decoder_Only/copilot/modeling_flax_llama.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6b3f1075d5cc01c4"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}