Learning to Refine: Self-Refinement of Parallel Reasoning in LLMs (2026.findings-acl)
Copied to clipboard
Qibin Wang, Pu Zhao, Shaohan Huang, Fangkai Yang, Lu Wang, Furu Wei, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang
| Challenge: | Existing approaches to test-time scaling are limited due to the quality of candidate responses. |
| Approach: | They propose a new metric to quantify the relative improvement of self-refinement beyond majority voting. |
| Outcome: | The proposed method achieves state-of-the-art performance across five benchmarks over other methods. |
Similar Papers
The ART of LLM Refinement: Ask, Refine, and Trust (2024.naacl-long)
Copied to clipboard
Kumar Shridhar, Koustuv Sinha, Andrew Cohen, Tianlu Wang, Ping Yu, Ramakanth Pasunuru, Mrinmaya Sachan, Jason Weston, Asli Celikyilmaz
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable generative abilities, but can they judge the quality of their own generations and self-improve? |
| Approach: | They propose a reasoning with a refinement strategy called *ART: Ask, Refine, and Trust* that asks necessary questions to decide when an LLM should refine its output and uses it to affirm or deny trust. |
| Outcome: | The proposed reasoning with a refinement strategy achieves a performance gain of +5 points over baselines on two multistep reasoning tasks. |
MAgICoRe: Multi-Agent, Iterative, Coarse-to-Fine Refinement for Reasoning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Excessive refinement can cause over-correction and reduce performance, authors say . they say MAgICoRe is a framework for multi-agent iteration for coarse-to-fine refinement . |
| Approach: | They propose a framework for multi-agent iteration for coarse-to-fine refinement that reduces excessive refinement by categorizing problems as easy or hard. |
| Outcome: | The proposed framework beats Self-Consistency by 3.4%, Best-of-k by 3.2%, and Self-Refine by 4.0% on Llama-3-8B and GPT- 3.5. |
Learning to Refine with Fine-Grained Natural Language Feedback (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work has explored the capability of large language models to identify and correct errors in LLM-generated responses. |
| Approach: | They propose to combine refinement with feedback into three distinct competencies . step 1: Detect, Critique, Refine gives a fine-grained feedback about errors . |
| Outcome: | The proposed method outperforms existing refinement approaches and models not fine-tuned for factuality critiquing. |
Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models (2025.emnlp-main)
Copied to clipboard
Kaiyan Chang, Yonghao Shi, Chenglong Wang, Hang Zhou, Chi Hu, Xiaoqian Liu, Yingfeng Luo, Yuan Ge, Tong Xiao, JingBo Zhu
| Challenge: | Recent training-based TTS methods, such as continued reinforcement learning, have surged in popularity, while training-free TTS approaches are gradually fading from prominence. |
| Approach: | They propose a fine-grained sequential scaling method guided by process verification that integrates training-free TTS methods with other classical parallel scaling methods at the step level. |
| Outcome: | Experiments on five instruction-tuned large language models (LLMs) show that training-free TTS methods can extend reasoning performance boundaries. |
Unlocking Recursive Thinking of LLMs: Alignment via Refinement (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for recursive reasoning are limited due to lack of expert-curated data. |
| Approach: | They propose a method that unlocks the potential of Large Language Models for recursive reasoning through long-form Chain of Thought. |
| Outcome: | The proposed method outperforms preference optimization methods on the openAI o1-series models by 20% on 3k synthetic samples. |
What Does LLM Refinement Actually Improve? A Systematic Study on Document-Level Literary Translation (2026.acl-long)
Copied to clipboard
Shaomu Tan, Dawei Zhu, Ke Tran, Michael Denkowski, Sony Trenous, Leonardo F. R. Ribeiro, Bill Byrne, Felix Hieber
| Challenge: | Large language models (LLMs) have made document-level machine translation increasingly practical, enabled by long-context modeling and strong generation quality. |
| Approach: | They propose to use document-level MT followed by segment-level refinement to find the strongest and most stable improvements across six LLMs and seven language pairs. |
| Outcome: | The proposed method outperforms error-specific prompting and evaluate-then-refine schemes in document-level translation. |
The Best of Both Worlds: Combining Parallel and Sequential Inference Scaling via Aggregation Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Empirical results show that AFT-trained models achieve substantial gains with test-time scaling. |
| Approach: | They introduce a supervised fine-tuning paradigm where models synthesize multiple draft responses into a single, refined answer. |
| Outcome: | Empirical results show that AFT-trained models outperform baseline models while eliminating external guidance. |
Parallel Test-Time Scaling for Latent Reasoning Models (2026.acl-long)
Copied to clipboard
| Challenge: | Parallel test-time scaling is a pivotal approach for enhancing large language models. |
| Approach: | They propose two uncertainty-inspired stochastic strategies for parallel test-time scaling for latent reasoning models and a Latent Reward Model for aggregation. |
| Outcome: | The proposed model scales well with compute and enables effective trajectory selection. |
Let’s Ask Again: Refine Network for Automatic Question Generation (D19-1)
Copied to clipboard
Preksha Nema, Akash Kumar Mohankumar, Mitesh M. Khapra, Balaji Vasan Srinivasan, Balaraman Ravindran
| Challenge: | Existing AQG models produce incomplete questions which look like incomplete drafts with scope for refinement. |
| Approach: | They propose a method which mimics the human process of generating questions by first creating an initial draft and then refining it. |
| Outcome: | The proposed method outperforms state-of-the-art methods on three datasets and improves on fluency and answerability metrics. |
Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to generate multiple independent CoTs, combining them through ensembling or other post-hoc strategies, have been shown to be effective in boosting performance. |
| Approach: | They propose a method where LLMs are fine-tuned to generate a sequence of Diverse Chains of Thought (DCoT) within a single inference step. |
| Outcome: | The proposed model can generate multiple chains of thought within a single inference step without external feedback. |