Lost in the Noise: How Reasoning Models Fail with Contextual Distractors

• URL: http://arxiv.org/abs/2601.07226v1
• Date: Mon, 12 Jan 2026 05:43:51 GMT
• Title: Lost in the Noise: How Reasoning Models Fail with Contextual Distractors
• Authors: Seongyun Lee, Yongrae Jo, Minju Seo, Moontae Lee, Minjoon Seo,
• Abstract summary: Recent advances in reasoning models and agentic AI systems have led to an increased reliance on diverse external information.<n>We introduce NoisyBench, a comprehensive benchmark that systematically evaluates model robustness across 11 datasets in RAG, reasoning, alignment, and tool-use tasks.<n>Our evaluation reveals a catastrophic performance drop of up to 80% in state-of-the-art models when faced with contextual distractors.
• Score: 57.31788955167306
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in reasoning models and agentic AI systems have led to an increased reliance on diverse external information. However, this shift introduces input contexts that are inherently noisy, a reality that current sanitized benchmarks fail to capture. We introduce NoisyBench, a comprehensive benchmark that systematically evaluates model robustness across 11 datasets in RAG, reasoning, alignment, and tool-use tasks against diverse noise types, including random documents, irrelevant chat histories, and hard negative distractors. Our evaluation reveals a catastrophic performance drop of up to 80% in state-of-the-art models when faced with contextual distractors. Crucially, we find that agentic workflows often amplify these errors by over-trusting noisy tool outputs, and distractors can trigger emergent misalignment even without adversarial intent. We find that prompting, context engineering, SFT, and outcome-reward only RL fail to ensure robustness; in contrast, our proposed Rationale-Aware Reward (RARE) significantly strengthens resilience by incentivizing the identification of helpful information within noise. Finally, we uncover an inverse scaling trend where increased test-time computation leads to worse performance in noisy settings and demonstrate via attention visualization that models disproportionately focus on distractor tokens, providing vital insights for building the next generation of robust, reasoning-capable agents.

Related papers

• AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Condition [72.24180896265192]
  We introduce AgentNoiseBench, a framework for evaluating robustness of agentic models under noisy environments.<n>We first conduct an in-depth analysis of biases and uncertainties in real-world scenarios.<n>We then categorize environmental noise into two primary types: user-noise and tool-noise.<n>Building on this analysis, we develop an automated pipeline that injects controllable noise into existing agent-centric benchmarks.
  arXiv  Detail & Related papers  (2026-02-11T20:33:10Z)
• RAGShaper: Eliciting Sophisticated Agentic RAG Skills via Automated Data Synthesis [29.39426376890088]
  Agentic Retrieval-Augmented Generation (RAG) empowers large language models to autonomously plan and retrieve information for complex problem-solving.<n>We introduce RAGShaper, a novel data synthesis framework designed to automate the construction of RAG tasks and robust agent trajectories.
  arXiv  Detail & Related papers  (2026-01-13T16:25:07Z)
• Self-Consistency as a Free Lunch: Reducing Hallucinations in Vision-Language Models via Self-Reflection [71.8243083897721]
  Vision-language models often hallucinate details, generating non-existent objects or inaccurate attributes that compromise output reliability.<n>We present a novel framework that leverages the model's self-consistency between long responses and short answers to generate preference pairs for training.
  arXiv  Detail & Related papers  (2025-09-27T10:37:11Z)
• From Entity Reliability to Clean Feedback: An Entity-Aware Denoising Framework Beyond Interaction-Level Signals [20.323837731778358]
  Implicit feedback is central to recommender systems but is inherently noisy, often impairing model training and degrading user experience.<n>We propose textbfEARD (textbfEntity-textbfAware textbfReliability-textbfDriven Denoising), a lightweight framework that shifts the focus from interaction-level signals to entity-level reliability.
  arXiv  Detail & Related papers  (2025-08-14T17:20:12Z)
• Automatically Identify and Rectify: Robust Deep Contrastive Multi-view Clustering in Noisy Scenarios [76.02688769599686]
  We propose a novel multi-view clustering framework for the automatic identification and rectification of noisy data, termed AIRMVC.<n>Specifically, we reformulate noisy identification as an anomaly identification problem using GMM.<n>We then design a hybrid rectification strategy to mitigate the adverse effects of noisy data based on the identification results.
  arXiv  Detail & Related papers  (2025-05-27T16:16:54Z)
• $C^2$AV-TSE: Context and Confidence-aware Audio Visual Target Speaker Extraction [80.57232374640911]
  We propose a model-agnostic strategy called the Mask-And-Recover (MAR)<n>MAR integrates both inter- and intra-modality contextual correlations to enable global inference within extraction modules.<n>To better target challenging parts within each sample, we introduce a Fine-grained Confidence Score (FCS) model.
  arXiv  Detail & Related papers  (2025-04-01T13:01:30Z)
• Adaptive Distraction: Probing LLM Contextual Robustness with Automated Tree Search [76.54475437069395]
  Large Language Models (LLMs) often struggle to maintain their original performance when faced with semantically coherent but task-irrelevant contextual information.<n>We propose a dynamic distraction generation framework based on tree search, where the generation process is guided by model behavior.
  arXiv  Detail & Related papers  (2025-02-03T18:43:36Z)
• RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response [23.45168175163634]
  Supervised fine-tuning (SFT) plays a crucial role in adapting large language models (LLMs) to specific domains or tasks.<n>We introduce a robust SFT framework (RobustFT) that performs noise detection and relabeling on downstream task data.<n>Extensive experiments conducted on multiple LLMs across five datasets demonstrate RobustFT's exceptional performance in noisy scenarios.
  arXiv  Detail & Related papers  (2024-12-19T15:00:18Z)
• Robust VAEs via Generating Process of Noise Augmented Data [9.366139389037489]
  This paper introduces a novel framework that enhances robustness by regularizing the latent space divergence between original and noise-augmented data. Our empirical evaluations demonstrate that this approach, termed Robust Augmented Variational Auto-ENcoder (RAVEN), yields superior performance in resisting adversarial inputs.
  arXiv  Detail & Related papers  (2024-07-26T09:55:34Z)
• $\text{R}^2$-Bench: Benchmarking the Robustness of Referring Perception Models under Perturbations [36.74309198908876]
  We present a comprehensive taxonomy of perturbations, and then develop a versatile toolbox for synthesizing and evaluating the effects of composite disturbances. We propose the $textR2$-Agent, an LLM-based agent that simplifies and automates model evaluation via natural language instructions.
  arXiv  Detail & Related papers  (2024-03-07T22:18:12Z)
• Improving a Named Entity Recognizer Trained on Noisy Data with a Few Clean Instances [55.37242480995541]
  We propose to denoise noisy NER data with guidance from a small set of clean instances. Along with the main NER model we train a discriminator model and use its outputs to recalibrate the sample weights. Results on public crowdsourcing and distant supervision datasets show that the proposed method can consistently improve performance with a small guidance set.
  arXiv  Detail & Related papers  (2023-10-25T17:23:37Z)
• CausalAgents: A Robustness Benchmark for Motion Forecasting using Causal Relationships [8.679073301435265]
  We construct a new benchmark for evaluating and improving model robustness by applying perturbations to existing data. We use these labels to perturb the data by deleting non-causal agents from the scene. Under non-causal perturbations, we observe a $25$-$38%$ relative change in minADE as compared to the original.
  arXiv  Detail & Related papers  (2022-07-07T21:28:23Z)

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.