BARREL: Boundary-Aware Reasoning for Factual and Reliable LRMs
yangjunxiao2021
Abstract
A novel framework, BARREL, addresses overconfidence in Large Reasoning Models by promoting concise and factual reasoning, significantly improving their reliability.
Recent advances in Large Reasoning Models (LRMs) have shown impressive capabilities in mathematical and logical reasoning. However, current LRMs rarely admit ignorance or respond with "I don't know". Instead, they often produce incorrect answers while showing undue confidence, raising concerns about their factual reliability. In this work, we identify two pathological reasoning patterns characterized by overthinking that contribute to the overconfident and incorrect answers: last-minute guessing and second-thought spiraling. To address these issues, we propose BARREL-a novel framework that promotes concise and boundary-aware factual reasoning. Our experiments show that BARREL-training increases the reliability of DeepSeek-R1-Distill-Llama-8B from 39.33% to 61.48%, while still achieving accuracy comparable to models finetuned on reasoning data generated by R1. These results demonstrate that our pilot study is inspiring to build more reliable and factual System 2 LRMs.
Community
Recent advances in Large Reasoning Models (LRMs) have shown impressive capabilities in mathematical and logical reasoning. However, current LRMs rarely admit ignorance or respond with "I don't know". Instead, they often produce incorrect answers while showing undue confidence, raising concerns about their factual reliability. In this work, we identify two pathological reasoning patterns characterized by overthinking that contribute to the overconfident and incorrect answers: last-minute guessing and second-thought spiraling. To address these issues, we propose BARREL-a novel framework that promotes concise and boundary-aware factual reasoning. Our experiments show that BARREL-training increases the reliability of DeepSeek-R1-Distill-Llama-8B from 39.33% to 61.48%, while still achieving accuracy comparable to models finetuned on reasoning data generated by R1. These results demonstrate that our pilot study is inspiring to build more reliable and factual System 2 LRMs.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- The Hallucination Tax of Reinforcement Finetuning (2025)
- RealSafe-R1: Safety-Aligned DeepSeek-R1 without Compromising Reasoning Capability (2025)
- SaRO: Enhancing LLM Safety through Reasoning-based Alignment (2025)
- Thinking Out Loud: Do Reasoning Models Know When They're Right? (2025)
- SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models (2025)
- Phi-4-Mini-Reasoning: Exploring the Limits of Small Reasoning Language Models in Math (2025)
- DRP: Distilled Reasoning Pruning with Skill-aware Step Decomposition for Efficient Large Reasoning Models (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper