Rethinking Supervised Fine-Tuning: Emphasizing Key Answer Tokens for Improved LLM Accuracy

• URL: http://arxiv.org/abs/2512.21017v1
• Date: Wed, 24 Dec 2025 07:24:31 GMT
• Title: Rethinking Supervised Fine-Tuning: Emphasizing Key Answer Tokens for Improved LLM Accuracy
• Authors: Xiaofeng Shi, Qian Kou, Yuduo Li, Hua Zhou,
• Abstract summary: Chain-of-Thought (CoT) component has become significant for complex reasoning tasks.<n>In conventional Supervised Fine-Tuning (SFT), the model could allocate disproportionately more attention to CoT sequences with excessive length.<n>We propose SFTKey, a two-stage training scheme. In the first stage, conventional SFT is applied to ensure proper output format, while in the second stage, only the Key portion is fine-tuned to improve accuracy.
• Score: 3.7208575749294392
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: With the rapid advancement of Large Language Models (LLMs), the Chain-of-Thought (CoT) component has become significant for complex reasoning tasks. However, in conventional Supervised Fine-Tuning (SFT), the model could allocate disproportionately more attention to CoT sequences with excessive length. This reduces focus on the much shorter but essential Key portion-the final answer, whose correctness directly determines task success and evaluation quality. To address this limitation, we propose SFTKey, a two-stage training scheme. In the first stage, conventional SFT is applied to ensure proper output format, while in the second stage, only the Key portion is fine-tuned to improve accuracy. Extensive experiments across multiple benchmarks and model families demonstrate that SFTKey achieves an average accuracy improvement exceeding 5\% over conventional SFT, while preserving the ability to generate correct formats. Overall, this study advances LLM fine-tuning by explicitly balancing CoT learning with additional optimization on answer-relevant tokens.

Related papers

• On-Policy Supervised Fine-Tuning for Efficient Reasoning [27.67711115864118]
  Large reasoning models (LRMs) are commonly trained with reinforcement learning (RL) to explore long chain-of-thought reasoning.<n>Recent methods add multi-reward objectives to jointly optimize correctness and brevity, but these complex extensions often destabilize training and yield suboptimal trade-offs.<n>We propose a simplified training strategy on-policy SFT, which reduces CoT length by up to 80 while maintaining original accuracy.
  arXiv  Detail & Related papers  (2026-02-13T19:16:39Z)
• CoT-Saliency: Unified Chain-of-Thought Reasoning for Heterogeneous Saliency Tasks [96.64597365827046]
  We present the first unified framework that jointly handles three operationally heterogeneous saliency tasks.<n>We introduce a Chain-of-Thought (CoT) reasoning process in a Vision-Language Model (VLM) to bridge task heterogeneity.<n>We show our model matches or outperforms specialized SOTA methods and strong closed-source VLMs across all tasks.
  arXiv  Detail & Related papers  (2025-11-01T04:37:01Z)
• Enhancing Large Language Model Reasoning via Selective Critical Token Fine-Tuning [18.934789236342244]
  Large language models (LLMs) primarily rely on supervised fine-tuning (SFT) to adapt pre-trained models to domain-specific tasks such as mathematical reasoning.<n>Standard SFT uniformly penalizes all tokens, neglecting that only a small subset of critical tokens determines reasoning correctness.<n>We propose Critical Token Fine-tuning (CFT), a simple yet effective approach that updates only tokens identified as functionally indispensable via counterfactual perturbations.
  arXiv  Detail & Related papers  (2025-10-13T03:25:36Z)
• Explore-Execute Chain: Towards an Efficient Structured Reasoning Paradigm [8.405729585427226]
  Chain-of-Thought (CoT) and its variants have markedly advanced the reasoning abilities of Large Language Models (LLMs)<n>We propose the Explore-Execute Chain ($E2C$), a structured reasoning framework that decouples reasoning into two distinct phases.
  arXiv  Detail & Related papers  (2025-09-28T15:48:40Z)
• On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectification [61.607788999847564]
  We present a simple yet theoretically motivated improvement to Supervised Fine-Tuning (SFT) for the Large Language Model (LLM)<n>We reveal that standard SFT gradients implicitly encode a problematic reward structure that may severely restrict the generalization capabilities of model.<n>We propose Dynamic Fine-Tuning (DFT), stabilizing gradient updates for each token by dynamically rescaling the objective function with the probability of this token.
  arXiv  Detail & Related papers  (2025-08-07T17:59:04Z)
• R-Stitch: Dynamic Trajectory Stitching for Efficient Reasoning [80.104336426172]
  Chain-of-thought (CoT) enhances problem-solving ability of large language models.<n>CoT incurs substantial inference cost due to long autoregressive trajectories.<n>We introduce R-Stitch, a training-free hybrid decoding framework.
  arXiv  Detail & Related papers  (2025-07-23T08:14:36Z)
• Implicit Reward as the Bridge: A Unified View of SFT and DPO Connections [65.36449542323277]
  We present a unified theoretical framework bridgingSupervised Fine-Tuning (SFT) and preference learning in Large Language Model (LLM) post-training.<n>We propose a simple yet effective learning rate reduction approach that yields significant performance improvements.
  arXiv  Detail & Related papers  (2025-06-15T05:42:29Z)
• Fractured Chain-of-Thought Reasoning [61.647243580650446]
  We introduce Fractured Sampling, a unified inference-time strategy that interpolates between full CoT and solution-only sampling.<n>We show that Fractured Sampling consistently achieves superior accuracy-cost trade-offs, yielding steep log-linear scaling gains in Pass@k versus token budget.
  arXiv  Detail & Related papers  (2025-05-19T11:30:41Z)
• Making Small Language Models Efficient Reasoners: Intervention, Supervision, Reinforcement [22.801244105119025]
  We propose new algorithms to improve token-efficient reasoning with small-scale models by effectively trading off accuracy and computation.<n>We first show that the post-SFT model fails to determine the optimal stopping point of the reasoning process, resulting in verbose and repetitive outputs.<n>Experiments on four reasoning benchmarks, MATH500, AMC, AIME24 and OlympiadBench, demonstrate that TS is highly effective compared to s1's budget forcing approach.
  arXiv  Detail & Related papers  (2025-05-12T18:04:39Z)
• The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models [69.798277882245]
  We introduce Unsupervised Prefix Fine-Tuning (UPFT) to enhance large language models' reasoning efficiency.<n>UPFT removes the need for labeled data or exhaustive sampling.<n> Experiments show that UPFT matches the performance of supervised methods.
  arXiv  Detail & Related papers  (2025-03-04T18:56:03Z)
• Quantize What Counts: More for Keys, Less for Values [63.51476878610841]
  Large Language Models (LLMs) suffer inference-time memory bottlenecks dominated by the attention Key-Value ( KV) cache.<n>This paper proposes two theorems that anchor mixed-precision KV quantization in the intrinsic geometry of Transformer models.
  arXiv  Detail & Related papers  (2025-02-20T22:24:27Z)
• Functional-level Uncertainty Quantification for Calibrated Fine-tuning on LLMs [38.33520071583312]
  Calibrated Fine-Tuning (UQ4CT) captures and calibrates uncertainty over the space of functions that map input prompts to outputs.<n>We implement UQ4CT during the fine-tuning stage via a mixture-of-experts framework that hierarchically decomposes the functional space.<n>Even under distribution shift, UQ4CT maintains superior ECE performance with high accuracy, showcasing improved generalizability.
  arXiv  Detail & Related papers  (2024-10-09T00:09:15Z)
• Intuitive Fine-Tuning: Towards Simplifying Alignment into a Single Process [19.986235452236272]
  Supervised Fine-Tuning (SFT) and Preference Optimization (PO) are key processes for aligning Language Models (LMs) with human preferences post pre-training.<n>We introduce Intuitive Fine-Tuning (IFT) to integrate SFT and PO into a single process.<n>IFT performs comparably or even superiorly to SFT and some typical PO methods across several tasks.
  arXiv  Detail & Related papers  (2024-05-20T08:23:28Z)
• Prefix Text as a Yarn: Eliciting Non-English Alignment in Foundation Language Model [50.339632513018934]
  supervised fine-tuning (SFT) has been a straightforward approach for tailoring the output of foundation large language model (LLM) to specific preferences. We critically examine this hypothesis within the scope of cross-lingual generation tasks. We introduce a novel training-free alignment method named PreTTY, which employs minimal task-related prior tokens.
  arXiv  Detail & Related papers  (2024-04-25T17:19:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.