ConflictScore: Identifying and Measuring How Language Models Handle Conflicting Evidence

• URL: http://arxiv.org/abs/2606.26437v1
• Date: Wed, 24 Jun 2026 23:00:09 GMT
• Title: ConflictScore: Identifying and Measuring How Language Models Handle Conflicting Evidence
• Abstract summary: We introduce ConflictScore, a novel metric that quantifies how well a model's response acknowledges conflicting evidence in its grounding documents.<n>Our framework decomposes responses into atomic claims, labels each claim against each grounding document, and then aggregates these labels into two complementary measures: ConflictScore-Count (CS-C), the proportion of claims exhibiting conflicts, and ConflictScore-Ratio (CS-R), the balance between supporting and contradicting evidence.
• Score: 48.938290419832335
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing metrics for factuality and faithfulness evaluate whether an answer is supported or contradicted by its grounding documents, but they fail to capture when both supporting and contradicting evidence coexist. We introduce ConflictScore, a novel metric that quantifies how well a model's response acknowledges conflicting evidence in its grounding documents. Our framework decomposes responses into atomic claims, labels each claim against each grounding document, and then aggregates these labels into two complementary measures: ConflictScore-Count (CS-C), the proportion of claims exhibiting conflicts, and ConflictScore-Ratio (CS-R), the balance between supporting and contradicting evidence. We develop ConflictBench, a benchmark covering diverse forms of conflicts such as ambiguity, contradiction, and divergent opinions, to systematically evaluate our metric. Experiments show that ConflictScore effectively detects overconfident claims across domains and can serve as a corrective feedback mechanism that improves truthfulness on TruthfulQA.

Related papers

• HiEviDR-Bench: A Benchmark for Hierarchical Evidence Aggregation in Deep Research [52.87128503345243]
  We introduce HiEviDR-Bench, a benchmark for evaluating Hierarchical Evidence Aggregation in Deep Research.<n>HiEviDR-Bench represents each instance with an explicit evidence graph that captures evidence selection, cross-source linking, and aggregation.<n>We develop a traceability-oriented evaluation framework with five dimensions: report quality, evidence traceability, citation accuracy, claim verification, and answer correctness.
  arXiv  Detail & Related papers  (2026-07-27T23:50:46Z)
• The Deliberative Illusion: Diagnosing Factual Attrition and Stance Homogenization in Multi-Agent LLM Deliberation [68.98390038721963]
  We show that multi-agent discussion erases up to 72% of issue-critical facts.<n>This loss is consequential: retained evidence can reconstruct the issue misleadingly, final stances remain anchored in base-model priors, and a single malicious agent can inject misinformation into the shrinking shared context.<n>We call for evaluations that measure which facts, uncertainties, and legitimate disagreements survive interaction.
  arXiv  Detail & Related papers  (2026-06-02T02:15:40Z)
• When Evidence Conflicts: Uncertainty and Order Effects in Retrieval-Augmented Biomedical Question Answering [2.6393907656700706]
  Using HealthContradict, we evaluate six open-weight large language models (LLMs) under five controlled evidence conditions.<n>Results show that conflicting biomedical evidence is both an uncertainty and robustness problem.
  arXiv  Detail & Related papers  (2026-05-13T21:02:24Z)
• When Reviews Disagree: Fine-Grained Contradiction Analysis in Scientific Peer Reviews [25.43097618275061]
  RevCI is an expert-annotated benchmark of peer-review pairs with evidence-level contradiction annotations with graded intensity labels.<n> IMPACT is a structured multi-agent framework that integrates aspect-conditioned evidence extraction, deliberative reasoning, and adjudication.<n>TIDE is a small language model that predicts contradiction evidence and intensity in a single forward pass.
  arXiv  Detail & Related papers  (2026-05-11T08:20:58Z)
• To Fuse or to Drop? Dual-Path Learning for Resolving Modality Conflicts in Multimodal Emotion Recognition [87.22988227382329]
  Multimodal emotion recognition (MER) benefits from combining text, audio, and vision, yet standard fusion often fails when modalities conflict.<n>We propose Dual-Path Conflict Resolution (DCR), a unified framework that learns when to fuse and when to drop modalities.
  arXiv  Detail & Related papers  (2026-05-06T13:11:33Z)
• Rethinking All Evidence: Enhancing Trustworthy Retrieval-Augmented Generation via Conflict-Driven Summarization [11.875601079871865]
  We propose CARE-RAG (Conflict-Aware and Reliable Evidence for RAG), a novel framework that improves trustworthiness through Conflict-Driven Summarization of all available evidence.<n>To detect and summarize conflicts, we distill a 3B LLaMA3.2 model to perform conflict-driven summarization, enabling reliable synthesis across multiple sources.<n>Experiments on revised QA datasets with retrieval data show that CARE-RAG consistently outperforms strong RAG baselines, especially in scenarios with noisy or conflicting evidence.
  arXiv  Detail & Related papers  (2025-07-02T01:39:49Z)
• SUCEA: Reasoning-Intensive Retrieval for Adversarial Fact-checking through Claim Decomposition and Editing [30.84752573088322]
  adversarial claims are intentionally designed by humans to challenge fact-checking systems.<n>We propose a training-free method designed to rephrase the original claim, making it easier to locate supporting evidence.<n>Our framework significantly improves on both retrieval and entailment label accuracy, outperforming four strong claim-decomposition-based baselines.
  arXiv  Detail & Related papers  (2025-06-05T02:58:15Z)
• Is the Top Still Spinning? Evaluating Subjectivity in Narrative Understanding [40.78154629252038]
  Forcing binary labels upon ambiguous claims lowers the reliability of evaluation.<n>We introduce LLM-generated edits of summaries as a method of providing a nuanced evaluation of claims.<n>We show that ARM produces a absolute 21% improvement in annotator agreement on claim faithfulness.
  arXiv  Detail & Related papers  (2025-04-01T19:08:24Z)
• AdaCAD: Adaptively Decoding to Balance Conflicts between Contextual and Parametric Knowledge [57.66282463340297]
  Knowledge conflict arises from discrepancies between information in the context of a large language model and the knowledge stored in its parameters.<n>We propose a fine-grained, instance-level approach called AdaCAD, which dynamically infers the weight of adjustment based on the degree of conflict.<n>We show that ADACAD consistently outperforms other decoding baselines with average QA accuracy gains of 14.21% (absolute) over a static contrastive baseline, and improves the factuality of summaries by 6.19 (AlignScore)
  arXiv  Detail & Related papers  (2024-09-11T16:35:18Z)
• End-to-End Page-Level Assessment of Handwritten Text Recognition [69.55992406968495]
  HTR systems increasingly face the end-to-end page-level transcription of a document. Standard metrics do not take into account the inconsistencies that might appear. We propose a two-fold evaluation, where the transcription accuracy and the RO goodness are considered separately.
  arXiv  Detail & Related papers  (2023-01-14T15:43:07Z)
• AmbiFC: Fact-Checking Ambiguous Claims with Evidence [57.7091560922174]
  We present AmbiFC, a fact-checking dataset with 10k claims derived from real-world information needs. We analyze disagreements arising from ambiguity when comparing claims against evidence in AmbiFC. We develop models for predicting veracity handling this ambiguity via soft labels.
  arXiv  Detail & Related papers  (2021-04-01T17:40:08Z)

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.