FindTheFlaws: Annotated Errors for Detecting Flawed Reasoning and Scalable Oversight Research

• URL: http://arxiv.org/abs/2503.22989v1
• Date: Sat, 29 Mar 2025 06:38:30 GMT
• Title: FindTheFlaws: Annotated Errors for Detecting Flawed Reasoning and Scalable Oversight Research
• Authors: Gabriel Recchia, Chatrik Singh Mangat, Issac Li, Gayatri Krishnakumar,
• Abstract summary: Approaches to scaling AI supervision include debate, critique, and prover-verifier games.<n>We present FindTheFlaws, a group of five diverse datasets spanning medicine, mathematics, science, coding, and the Lojban language.<n>We evaluate frontier models' critiquing capabilities and observe a range of performance that can be leveraged for scalable oversight experiments.
• Score: 0.6286531904189063
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As AI models tackle increasingly complex problems, ensuring reliable human oversight becomes more challenging due to the difficulty of verifying solutions. Approaches to scaling AI supervision include debate, in which two agents engage in structured dialogue to help a judge evaluate claims; critique, in which models identify potential flaws in proposed solutions; and prover-verifier games, in which a capable 'prover' model generates solutions that must be verifiable by a less capable 'verifier'. Evaluations of the scalability of these and similar approaches to difficult problems benefit from datasets that include (1) long-form expert-verified correct solutions and (2) long-form flawed solutions with annotations highlighting specific errors, but few are available. To address this gap, we present FindTheFlaws, a group of five diverse datasets spanning medicine, mathematics, science, coding, and the Lojban language. Each dataset contains questions and long-form solutions with expert annotations validating their correctness or identifying specific error(s) in the reasoning. We evaluate frontier models' critiquing capabilities and observe a range of performance that can be leveraged for scalable oversight experiments: models performing more poorly on particular datasets can serve as judges/verifiers for more capable models. Additionally, for some task/dataset combinations, expert baselines exceed even top model performance, making them more beneficial for scalable oversight experiments.

Related papers

• Adaptively evaluating models with task elicitation [29.704450391533864]
  We introduce and validate a framework for evaluating language models called Adaptive Evaluations. We find that frontier models lack consistency when adaptively probed with our framework on a diverse suite of datasets and tasks. Generated questions pass human validity checks and often transfer to other models with different capability profiles.
  arXiv  Detail & Related papers  (2025-03-03T19:04:10Z)
• From Correctness to Comprehension: AI Agents for Personalized Error Diagnosis in Education [24.970741456147447]
  Large Language Models (LLMs) have demonstrated impressive mathematical reasoning capabilities, achieving near-perfect performance on benchmarks like GSM8K. However, their application in personalized education remains limited due to an overemphasis on correctness over error diagnosis and feedback generation. We introduce textbfMathCCS, a benchmark designed for systematic error analysis and tailored feedback. Second, we develop a sequential error analysis framework that leverages historical data to track trends and improve diagnostic precision. Third, we propose a multi-agent collaborative framework that combines a Time Series Agent for historical analysis and an MLLM Agent for real-
  arXiv  Detail & Related papers  (2025-02-19T14:57:51Z)
• EnigmaEval: A Benchmark of Long Multimodal Reasoning Challenges [17.056693711040747]
  We introduce EnigmaEval, a dataset of problems and solutions derived from puzzle competitions and events.<n>This dataset probes models' ability to perform implicit knowledge synthesis and multi-step deductive reasoning.<n>The benchmark comprises 1184 puzzles of varying complexity requiring teams of skilled solvers hours to days to complete.
  arXiv  Detail & Related papers  (2025-02-13T00:18:34Z)
• Subtle Errors Matter: Preference Learning via Error-injected Self-editing [59.405145971637204]
  We propose a novel preference learning framework called eRror-Injected Self-Editing (RISE)<n>RISE injects predefined subtle errors into pivotal tokens in reasoning or steps to construct hard pairs for error mitigation.<n>Experiments validate the effectiveness of RISE, with preference learning on Qwen2-7B-Instruct yielding notable improvements of 3.0% on GSM8K and 7.9% on MATH with only 4.5K training samples.
  arXiv  Detail & Related papers  (2024-10-09T07:43:38Z)
• Unsupervised Model Diagnosis [49.36194740479798]
  This paper proposes Unsupervised Model Diagnosis (UMO) to produce semantic counterfactual explanations without any user guidance. Our approach identifies and visualizes changes in semantics, and then matches these changes to attributes from wide-ranging text sources.
  arXiv  Detail & Related papers  (2024-10-08T17:59:03Z)
• ErrorRadar: Benchmarking Complex Mathematical Reasoning of Multimodal Large Language Models Via Error Detection [60.297079601066784]
  We introduce ErrorRadar, the first benchmark designed to assess MLLMs' capabilities in error detection. ErrorRadar evaluates two sub-tasks: error step identification and error categorization. It consists of 2,500 high-quality multimodal K-12 mathematical problems, collected from real-world student interactions. Results indicate significant challenges still remain, as GPT-4o with best performance is still around 10% behind human evaluation.
  arXiv  Detail & Related papers  (2024-10-06T14:59:09Z)
• Can Large Language Models Always Solve Easy Problems if They Can Solve Harder Ones? [65.43882564649721]
  Large language models (LLMs) have demonstrated impressive capabilities, but still suffer from inconsistency issues. We develop the ConsisEval benchmark, where each entry comprises a pair of questions with a strict order of difficulty. We analyze the potential for improvement in consistency by relative consistency score.
  arXiv  Detail & Related papers  (2024-06-18T17:25:47Z)
• Optimizing Language Model's Reasoning Abilities with Weak Supervision [48.60598455782159]
  We present textscPuzzleBen, a weakly supervised benchmark that comprises 25,147 complex questions, answers, and human-generated rationales. A unique aspect of our dataset is the inclusion of 10,000 unannotated questions, enabling us to explore utilizing fewer supersized data to boost LLMs' inference capabilities.
  arXiv  Detail & Related papers  (2024-05-07T07:39:15Z)
• A Reliable Framework for Human-in-the-Loop Anomaly Detection in Time Series [17.08674819906415]
  We introduce HILAD, a novel framework designed to foster a dynamic and bidirectional collaboration between humans and AI. Through our visual interface, HILAD empowers domain experts to detect, interpret, and correct unexpected model behaviors at scale.
  arXiv  Detail & Related papers  (2024-05-06T07:44:07Z)
• Exploring the Trade-off between Plausibility, Change Intensity and Adversarial Power in Counterfactual Explanations using Multi-objective Optimization [73.89239820192894]
  We argue that automated counterfactual generation should regard several aspects of the produced adversarial instances. We present a novel framework for the generation of counterfactual examples.
  arXiv  Detail & Related papers  (2022-05-20T15:02:53Z)
• Generalization of Neural Combinatorial Solvers Through the Lens of Adversarial Robustness [68.97830259849086]
  Most datasets only capture a simpler subproblem and likely suffer from spurious features. We study adversarial robustness - a local generalization property - to reveal hard, model-specific instances and spurious features. Unlike in other applications, where perturbation models are designed around subjective notions of imperceptibility, our perturbation models are efficient and sound. Surprisingly, with such perturbations, a sufficiently expressive neural solver does not suffer from the limitations of the accuracy-robustness trade-off common in supervised learning.
  arXiv  Detail & Related papers  (2021-10-21T07:28:11Z)
• Discovering and Validating AI Errors With Crowdsourced Failure Reports [10.4818618376202]
  We introduce crowdsourced failure reports, end-user descriptions of how or why a model failed, and show how developers can use them to detect AI errors. We also design and implement Deblinder, a visual analytics system for synthesizing failure reports. In semi-structured interviews and think-aloud studies with 10 AI practitioners, we explore the affordances of the Deblinder system and the applicability of failure reports in real-world settings.
  arXiv  Detail & Related papers  (2021-09-23T23:26:59Z)

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.