Lune

EMNLP Findings2024

Self-contradictory reasoning evaluation and detection

Ziyi Liu, Soumya Sanyal, Isabelle Lee, Yongkang Du, Rahul Gupta, Yang Liu, Jieyu Zhao

Abstract

In a plethora of recent work, large language models (LLMs) demonstrated impressive reasoning ability, but many proposed downstream reasoning tasks only focus on final answers. Two fundamental questions persist: 1) how consistent is the reasoning, and 2) can models detect unreliable reasoning? In this paper, we investigate self-contradictory (SELF-CONTRA) reasoning, where the model reasoning does not support its answers. To answer 1), we define and assess the SELF-CONTRA rate across three datasets and delve into finer-grained categories of SELF-CONTRA reasoning. We find that LLMs often contradict themselves in reasoning tasks involving contextual information understanding or commonsense. The model may generate correct answers by taking shortcuts in reasoning or overlooking contextual evidence, leading to compromised reasoning. For 2), we task the state-of-the-art model GPT-4 with identifying SELF-CONTRA reasoning and finergrained fallacies. We find that finer-grained categories enhanced detection can improve GPT-4's ability to detect SELF-CONTRA. However, it is only able to detect SELF-CONTRA with a 52.2% F1 score, much lower compared to 66.7% for humans. Our results indicate that current LLMs lack the robustness necessary for reliable reasoning and we emphasize the urgent need for establishing best practices in comprehensive reasoning evaluations beyond pure performance-based metrics. 1

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

4 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Large Language Models Cannot Self-Correct Reasoning Yet cites Large Language Models are Zero-Shot Reasoners
  • Large Language Models Cannot Self-Correct Reasoning Yet cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Enhancing Reasoning Capabilities of LLMs via Principled Synthetic Logic Corpus cites Large Language Models Cannot Self-Correct Reasoning Yet
  • Enhancing Reasoning Capabilities of LLMs via Principled Synthetic Logic Corpus cites Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting
  • Enhancing Reasoning Capabilities of LLMs via Principled Synthetic Logic Corpus cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting cites Large Language Models are Zero-Shot Reasoners
  • Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Uncertainty in Causality: A New Frontier cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  • Large Language Models are Zero-Shot Reasoners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers4

  1. Enhancing Reasoning Capabilities of LLMs via Principled Synthetic Logic CorpusTerufumi Morishita, Gaku Morio, Atsuki Yamaguchi, Yasuhiro SogawaNeurIPS 2024 · 60 citations
  2. Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination DetectionYongxin Deng, Zhen Fang, Sharon Li, Ling ChenICLR 2026 · 5 citations
  3. Uncertainty in Causality: A New FrontierShaobo Cui, Luca Mouchel, Boi FaltingsACL 2025
  4. The Consistency Dilemma in LLMs: Generator-Evaluator Agreement and Vulnerability to MistakesMarina Mancoridis, Zoe HitzigICML 2026

Builds on19

  1. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
  2. Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations
  3. Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought PromptingMiles Turpin, Julian Michael, Ethan Perez, Samuel R. BowmanNeurIPS 2023 · 1,792 citations
  4. ChatEval: Towards Better LLM-based Evaluators through Multi-Agent DebateChi-Min Chan, Weize Chen, Yusheng Su, Jianxuan Yu et al.ICLR 2024 · 871 citations
  5. Large Language Models Cannot Self-Correct Reasoning YetJie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng et al.ICLR 2024 · 858 citations

Related papers

  1. Calibrating Reasoning in Language Models with Internal ConsistencyZhihui Xie, Jizhou Guo, Tong Yu, Shuai LiNeurIPS 2024 · 37 citations
  2. Conditional and Modal Reasoning in Large Language ModelsWesley H. Holliday, Matthew Mandelkern, Cedegao ZhangEMNLP 2024 · 5 citations
  3. Exposing the Achilles' Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical ReasoningJoykirat Singh, Akshay Uttama Nambi, Vibhav VineetACL 2025 · 10 citations
  4. Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones?Zhe Yang, Yichang Zhang, Tianyu Liu, Jian Yang et al.EMNLP 2024 · 2 citations
  5. Benchmarking LLM's Capability in Reasoning over Conflicting Web ReferencesYizhen Yuan, Rui Kong, Dongze Li, Yuanchun Li et al.ACL 2026
Dusk over the sea between two cliffs drawn in fine vertical lines