Bias-Restrained Prefix Representation Finetuning for Mathematical Reasoning

• URL: http://arxiv.org/abs/2511.10707v1
• Date: Thu, 13 Nov 2025 05:15:36 GMT
• Title: Bias-Restrained Prefix Representation Finetuning for Mathematical Reasoning
• Authors: Sirui Liang, Pengfei Cao, Jian Zhao, Cong Huang, Jun Zhao, Kang Liu,
• Abstract summary: Representation finetuning (ReFT) methods improve efficiency by freezing model weights and optimizing internal representations with fewer parameters than PEFT.<n>ReFT exhibits a significant performance decline on mathematical reasoning tasks.<n>This paper proposes Bias-REstrained Prefix Representation FineTuning (BREP ReFT), which enhances ReFT's mathematical reasoning capability.
• Score: 33.28699044085956
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Parameter-Efficient finetuning (PEFT) enhances model performance on downstream tasks by updating a minimal subset of parameters. Representation finetuning (ReFT) methods further improve efficiency by freezing model weights and optimizing internal representations with fewer parameters than PEFT, outperforming PEFT on several tasks. However, ReFT exhibits a significant performance decline on mathematical reasoning tasks. To address this problem, the paper demonstrates that ReFT's poor performance on mathematical tasks primarily stems from its struggle to generate effective reasoning prefixes during the early inference phase. Moreover, ReFT disturbs the numerical encoding and the error accumulats during the CoT stage. Based on these observations, this paper proposes Bias-REstrained Prefix Representation FineTuning (BREP ReFT), which enhances ReFT's mathematical reasoning capability by truncating training data to optimize the generation of initial reasoning prefixes, intervening on the early inference stage to prevent error accumulation, and constraining the intervention vectors' magnitude to avoid disturbing numerical encoding. Extensive experiments across diverse model architectures demonstrate BREP's superior effectiveness, efficiency, and robust generalization capability, outperforming both standard ReFT and weight-based PEFT methods on the task of mathematical reasoning. The source code is available at https://github.com/LiangThree/BREP.

Related papers

• Benchmarking Few-shot Transferability of Pre-trained Models with Improved Evaluation Protocols [123.73663884421272]
  Few-shot transfer has been revolutionized by stronger pre-trained models and improved adaptation algorithms.<n>We establish FEWTRANS, a comprehensive benchmark containing 10 diverse datasets.<n>By releasing FEWTRANS, we aim to provide a rigorous "ruler" to streamline reproducible advances in few-shot transfer learning research.
  arXiv  Detail & Related papers  (2026-02-28T05:41:57Z)
• Gradually Compacting Large Language Models for Reasoning Like a Boiling Frog [72.4168434368873]
  Large Language Models (LLMs) have demonstrated impressive reasoning capabilities, but their substantial size often demands significant computational resources.<n>We propose a gradual compacting method that divides the compression process into multiple fine-grained iterations.<n>This iterative approach-reminiscent of the "boiling frog" effect-enables the model to be progressively compressed without abrupt performance loss.
  arXiv  Detail & Related papers  (2026-02-04T06:56:52Z)
• High-Rank Structured Modulation for Parameter-Efficient Fine-Tuning [57.85676271833619]
  Low-rank Adaptation (LoRA) uses a low-rank update method to simulate full parameter fine-tuning.<n>We present textbfSMoA, a high-rank textbfStructured textbfMOdulation textbfAdapter that uses fewer trainable parameters while maintaining a higher rank.
  arXiv  Detail & Related papers  (2026-01-12T13:06:17Z)
• Task-Adaptive Parameter-Efficient Fine-Tuning for Weather Foundation Models [28.79157031663951]
  We introduce WeatherPEFT, a novel PEFT framework for Weather Foundation Models (WFMs)<n>Task-Tu Dynamic Prompting (TTu) injects the embedding weights within the encoder to the input tokens of the backbone of the pre-trained via internal and external pattern extraction.<n>Fisher-Guided Adaptive Selection (SFAS) identifies the most task-critical randomness, preserving in pre-trained knowledge, but also stabilizing the selection.<n>We demonstrate the effectiveness and efficiency of WeatherPEFT on three downstream tasks, where existing PEFT methods show significant gaps versus Full-Tuning, and WeatherPEFT performance parity
  arXiv  Detail & Related papers  (2025-09-26T07:54:05Z)
• Nested-ReFT: Efficient Reinforcement Learning for Large Language Model Fine-Tuning via Off-Policy Rollouts [25.205293698698867]
  We introduce Nested-ReFT, where a subset of layers of the target model acts as the behavior model to generate off-policy completions during training.<n>Our theoretical analysis shows that Nested-ReFT yields unbiased gradient estimates with controlled variance.<n>Our empirical analysis demonstrates improved computational efficiency measured as tokens/sec across multiple math reasoning benchmarks and model sizes.
  arXiv  Detail & Related papers  (2025-08-13T18:37:46Z)
• Enhancing Chain-of-Thought Reasoning with Critical Representation Fine-tuning [37.16998362490576]
  Representation Fine-tuning (ReFT) has attracted widespread attention for significantly improving parameter efficiency by editing representation space alone.<n>We propose Critical Representation Fine-Tuning (CRFT), a novel method that identifies and optimize these critical representations through information flow analysis.<n>Our method is validated across eight benchmarks for arithmetic and commonsense reasoning, using LLaMA and Mistral model families.
  arXiv  Detail & Related papers  (2025-07-14T09:11:33Z)
• PiCa: Parameter-Efficient Fine-Tuning with Column Space Projection [9.152388341742048]
  Fine-tuning large foundation models is essential for building expert models tailored to specialized tasks and domains.<n>We introduce efficient Fine-tuning with Column Space Projection (PiCa), a novel theoretically grounded PEFT method.<n>PiCa consistently outperforms state-of-the-art baselines under comparable or smaller parameter budgets.
  arXiv  Detail & Related papers  (2025-05-26T16:52:40Z)
• NEAT: Nonlinear Parameter-efficient Adaptation of Pre-trained Models [26.808251361020066]
  Fine-tuning pre-trained models often yields state-of-the-art performance but is computationally expensive when updating all parameters.<n>We propose NEAT, a nonlinear PEFT approach that employs a lightweight neural network to learn a nonlinear transformation of the pre-trained weights.<n>Our theoretical analysis shows that NEAT achieves greater efficiency than LoRA while maintaining equivalent expressivity.
  arXiv  Detail & Related papers  (2024-10-02T17:29:23Z)
• Forecast-PEFT: Parameter-Efficient Fine-Tuning for Pre-trained Motion Forecasting Models [68.23649978697027]
  Forecast-PEFT is a fine-tuning strategy that freezes the majority of the model's parameters, focusing adjustments on newly introduced prompts and adapters. Our experiments show that Forecast-PEFT outperforms traditional full fine-tuning methods in motion prediction tasks. Forecast-FT further improves prediction performance, evidencing up to a 9.6% enhancement over conventional baseline methods.
  arXiv  Detail & Related papers  (2024-07-28T19:18:59Z)
• Sparse is Enough in Fine-tuning Pre-trained Large Language Models [98.46493578509039]
  We propose a gradient-based sparse fine-tuning algorithm, named Sparse Increment Fine-Tuning (SIFT) We validate its effectiveness on a range of tasks including the GLUE Benchmark and Instruction-tuning.
  arXiv  Detail & Related papers  (2023-12-19T06:06:30Z)
• Boosting Inference Efficiency: Unleashing the Power of Parameter-Shared Pre-trained Language Models [109.06052781040916]
  We introduce a technique to enhance the inference efficiency of parameter-shared language models. We also propose a simple pre-training technique that leads to fully or partially shared models. Results demonstrate the effectiveness of our methods on both autoregressive and autoencoding PLMs.
  arXiv  Detail & Related papers  (2023-10-19T15:13:58Z)
• LoRAPrune: Structured Pruning Meets Low-Rank Parameter-Efficient Fine-Tuning [56.88751562302793]
  Low-rank adaption (LoRA) has emerged to fine-tune large language models (LLMs) LoRAPrune is a new framework that delivers an accurate structured pruned model in a highly memory-efficient manner. LoRAPrune achieves a reduction in perplexity by 4.81 on WikiText2 and 3.46 on PTB, while also decreasing memory usage by 52.6%.
  arXiv  Detail & Related papers  (2023-05-28T15:15:48Z)

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.