paper-with-me

홈 › Papers

LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints

2024-10-09 · Thomas Palmeira Ferraz, Kartik Mehta, Yu-Hsiang Lin, Haw-Shiuan Chang, Shereen Oraby, Sijia Liu, Vivek Subramanian, Tagyoung Chung, Mohit Bansal, Nanyun Peng

Instruction following is a key capability for LLMs. However, recent studies have shown that LLMs often struggle with instructions containing multiple constraints (e.g. a request to create a social media post "in a funny tone" with "no hashtag"). Despite this, most evaluations focus solely on synthetic data. To address this, we introduce RealInstruct, the first benchmark designed to evaluate LLMs' ability to follow real-world multi-constrained instructions by leveraging queries real users asked AI assistants. We also investigate model-based evaluation as a cost-effective alternative to human annotation for this task. Our findings reveal that even the proprietary GPT-4 model fails to meet at least one constraint on over 21% of instructions, highlighting the limitations of state-of-the-art models. To address the performance gap between open-source and proprietary models, we propose the Decompose, Critique and Refine (DeCRIM) self-correction pipeline, which enhances LLMs' ability to follow constraints. DeCRIM works by decomposing the original instruction into a list of constraints and using a Critic model to decide when and where the LLM's response needs refinement. Our results show that DeCRIM improves Mistral's performance by 7.3% on RealInstruct and 8.0% on IFEval even with weak feedback. Moreover, we demonstrate that with strong feedback, open-source LLMs with DeCRIM can outperform GPT-4 on both benchmarks.

📄 PDF Abstract BibTeX arXiv:2410.06458

Code (0)

등록된 구현이 없습니다.

Tasks

Instruction Following

Methods 이 논문이 사용한 방법론

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Adam 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Attention 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs

2024-12-27 · Zhe Yang, Yichang Zhang, Yudong Wang, Ziyao Xu 외

Large Language Models (LLMs) can correct their self-generated responses, but a decline in accuracy after self-correction is also witnessed. To have a deeper understanding of self-correction, we endeavor to decompose, eva…

In-Context Learning

Small Language Models Need Strong Verifiers to Self-Correct Reasoning

2024-04-26 · Yunxiang Zhang, Muhammad Khalifa, Lajanugen Logeswaran, Jaekyeom Kim 외

Self-correction has emerged as a promising solution to boost the reasoning performance of large language models (LLMs), where LLMs refine their solutions using self-generated critiques that pinpoint the errors. This work…

Math

MCQG-SRefine: Multiple Choice Question Generation and Evaluation with Iterative Self-Critique, Correction, and Comparison Feedback

2024-10-17 · Zonghai Yao, Aditya Parashar, Huixue Zhou, Won Seok Jang 외

Automatic question generation (QG) is essential for AI and NLP, particularly in intelligent tutoring, dialogue systems, and fact verification. Generating multiple-choice questions (MCQG) for professional exams, like the …

Fact VerificationHallucinationMultiple-choicePrompt Engineering+2

CriticBench: Benchmarking LLMs for Critique-Correct Reasoning

2024-02-22 · Zicheng Lin, Zhibin Gou, Tian Liang, Ruilin Luo 외

The ability of Large Language Models (LLMs) to critique and refine their reasoning is crucial for their application in evaluation, feedback provision, and self-improvement. This paper introduces CriticBench, a comprehens…

Benchmarking

Beyond Output Critique: Self-Correction via Task Distillation

2026-01-31 · Hossein A. Rahmani, Mengting Wan, Pei Zhou, Longqi Yang 외 arxiv

Large language models (LLMs) have shown promising self-correction abilities, where iterative refinement improves the quality of generated responses. However, most existing approaches operate at the level of output critiq…