Firm or Fickle? Evaluating Large Language Models Consistency in Sequential Interactions

• URL: http://arxiv.org/abs/2503.22353v1
• Date: Fri, 28 Mar 2025 11:49:56 GMT
• Title: Firm or Fickle? Evaluating Large Language Models Consistency in Sequential Interactions
• Authors: Yubo Li, Yidi Miao, Xueying Ding, Ramayya Krishnan, Rema Padman,
• Abstract summary: Large Language Models (LLMs) have shown remarkable capabilities across various tasks, but their deployment in high-stake domains requires consistent performance across multiple interaction rounds.<n>This paper introduces a comprehensive framework for evaluating and improving LLM response consistency, making three key contributions.
• Score: 8.069858557211132
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Language Models (LLMs) have shown remarkable capabilities across various tasks, but their deployment in high-stake domains requires consistent performance across multiple interaction rounds. This paper introduces a comprehensive framework for evaluating and improving LLM response consistency, making three key contributions. First, we propose a novel Position-Weighted Consistency (PWC) score that captures both the importance of early-stage stability and recovery patterns in multi-turn interactions. Second, we present a carefully curated benchmark dataset spanning diverse domains and difficulty levels, specifically designed to evaluate LLM consistency under various challenging follow-up scenarios. Third, we introduce Confidence-Aware Response Generation (CARG), a framework that significantly improves response stability by incorporating model confidence signals into the generation process. Empirical results demonstrate that CARG significantly improves response stability without sacrificing accuracy, underscoring its potential for reliable LLM deployment in critical applications.

Related papers

• A Survey on Post-training of Large Language Models [185.51013463503946]
  Large Language Models (LLMs) have fundamentally transformed natural language processing, making them indispensable across domains ranging from conversational systems to scientific exploration.<n>These challenges necessitate advanced post-training language models (PoLMs) to address shortcomings, such as restricted reasoning capacities, ethical uncertainties, and suboptimal domain-specific performance.<n>This paper presents the first comprehensive survey of PoLMs, systematically tracing their evolution across five core paradigms.
  arXiv  Detail & Related papers  (2025-03-08T05:41:42Z)
• Collective Reasoning Among LLMs A Framework for Answer Validation Without Ground Truth [0.0]
  This study explores how inter-model consensus enhances response reliability and serves as a proxy for assessing the quality of generated questions. We present a collaborative framework where multiple large language models, namely GPT-4-0125-preview, Meta-LLaMA-3-70B-Instruct, Claude-3-Opus, and Gemini-1.5-Flash, work together to generate and respond to complex PhD-level probability questions.
  arXiv  Detail & Related papers  (2025-02-28T06:20:52Z)
• Progressive Multimodal Reasoning via Active Retrieval [64.74746997923967]
  Multi-step multimodal reasoning tasks pose significant challenges for large language models (MLLMs) We propose AR-MCTS, a universal framework designed to progressively improve the reasoning capabilities of MLLMs. We show that AR-MCTS can optimize sampling diversity and accuracy, yielding reliable multimodal reasoning.
  arXiv  Detail & Related papers  (2024-12-19T13:25:39Z)
• Are Your LLMs Capable of Stable Reasoning? [38.03049704515947]
  Large Language Models (LLMs) have demonstrated remarkable progress in complex reasoning tasks.<n>However, a significant discrepancy persists between benchmark performances and real-world applications.<n>We introduce G-Pass@k, a novel evaluation metric that provides a continuous assessment of model performance.<n>We present LiveMathBench, a dynamic benchmark comprising challenging, contemporary mathematical problems.
  arXiv  Detail & Related papers  (2024-12-17T18:12:47Z)
• On Adversarial Robustness and Out-of-Distribution Robustness of Large Language Models [0.16874375111244325]
  We investigate the correlation between adversarial robustness and OOD robustness in large language models (LLMs) Our findings highlight nuanced interactions between adversarial robustness and OOD robustness, with results indicating limited transferability. Further research is needed to evaluate these interactions across larger models and varied architectures.
  arXiv  Detail & Related papers  (2024-12-13T20:04:25Z)
• Evaluating and Advancing Multimodal Large Language Models in Ability Lens [30.083110119139793]
  We introduce textbfAbilityLens, a unified benchmark designed to evaluate MLLMs across six key perception abilities. We identify the strengths and weaknesses of current models, highlighting stability patterns and revealing a notable performance gap between open-source and closed-source models. We also design a simple ability-specific model merging method that combines the best ability checkpoint from early training stages, effectively mitigating performance decline due to ability conflict.
  arXiv  Detail & Related papers  (2024-11-22T04:41:20Z)
• Reward-Robust RLHF in LLMs [25.31456438114974]
  Large Language Models (LLMs) continue to progress toward more advanced forms of intelligence. The reliance on reward-model-based (RM-based) alignment methods introduces significant challenges. We introduce a reward-robust RLHF framework aimed at addressing these fundamental challenges.
  arXiv  Detail & Related papers  (2024-09-18T02:35:41Z)
• MARS: Benchmarking the Metaphysical Reasoning Abilities of Language Models with a Multi-task Evaluation Dataset [50.36095192314595]
  Large Language Models (LLMs) function as conscious agents with generalizable reasoning capabilities. This ability remains underexplored due to the complexity of modeling infinite possible changes in an event. We introduce the first-ever benchmark, MARS, comprising three tasks corresponding to each step.
  arXiv  Detail & Related papers  (2024-06-04T08:35:04Z)
• Measuring and Improving Chain-of-Thought Reasoning in Vision-Language Models [61.28463542324576]
  Vision-language models (VLMs) have recently demonstrated strong efficacy as visual assistants that can generate human-like outputs. We evaluate existing state-of-the-art VLMs and find that even the best-performing model is unable to demonstrate strong visual reasoning capabilities and consistency. We propose a two-stage training framework aimed at improving both the reasoning performance and consistency of VLMs.
  arXiv  Detail & Related papers  (2023-09-08T17:49:44Z)
• Improving Open Information Extraction with Large Language Models: A Study on Demonstration Uncertainty [52.72790059506241]
  Open Information Extraction (OIE) task aims at extracting structured facts from unstructured text. Despite the potential of large language models (LLMs) like ChatGPT as a general task solver, they lag behind state-of-the-art (supervised) methods in OIE tasks.
  arXiv  Detail & Related papers  (2023-09-07T01:35:24Z)
• Two Failures of Self-Consistency in the Multi-Step Reasoning of LLMs [78.31625291513589]
  We argue that self-consistency is an important criteria for valid multi-step reasoning in tasks where the solution is composed of the answers to multiple sub-steps. We propose two types of self-consistency that are particularly important for multi-step reasoning -- hypothetical consistency and compositional consistency. We demonstrate that multiple variants of the GPT-3/-4 models exhibit poor consistency rates across both types of consistency on a variety of tasks.
  arXiv  Detail & Related papers  (2023-05-23T17:25:59Z)

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.