Order Matters: LVLMs as Judges for Temporal Reasoning in Image Sequences

• URL: http://arxiv.org/abs/2608.10908v1
• Date: Tue, 11 Aug 2026 13:29:03 GMT
• Title: Order Matters: LVLMs as Judges for Temporal Reasoning in Image Sequences
• Abstract summary: We argue that the reliance on Large Vision-Language Models as judges is fundamentally limited by architectural biases.<n>Our analysis reveals a profound performance dichotomy: while models appear competent in isolated pointwise scoring, they suffer a catastrophic collapse when required to perform pairwise discrimination of temporal order.<n>These biases, potentially rooted in causal masking and rotary embeddings, suggest that current transformer-based judges are inherently ill-equipped for long-form visual reasoning.
• Score: 0.9841060883971746
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: As generative multimedia evolves from static image synthesis to complex, interleaved visual narratives, a foundational bottleneck has emerged: the judgment crisis. While human perception naturally synthesizes the temporal and logical flow of a story, automated evaluation systems remain largely "blind" to sequential continuity, often failing to distinguish between a coherent narrative and a semantically shuffled or contradictory sequence. This work identifies a critical structural gap in current multimodal evaluation paradigms, arguing that the reliance on Large Vision-Language Models (LVLMs) as judges is fundamentally limited by architectural biases. Our analysis reveals a profound performance dichotomy: while models may appear competent in isolated pointwise scoring, they suffer a catastrophic collapse when required to perform pairwise discrimination of temporal order. We demonstrate that this is not merely a data-scarcity issue but a structural one. Through a series of diagnostic probes, we uncover systematic positional asymmetries, specifically primacy and recency effects, where a model's judgment of a story is significantly influenced by the placement of a frame, often more than by its semantic consistency. These biases, potentially rooted in causal masking and rotary embeddings, suggest that current transformer-based judges are inherently ill-equipped for long-form visual reasoning. By exposing these blind spots, we challenge the multimedia community to move beyond snapshot-centric metrics and instead pioneer Temporally-Aware Evaluation paradigms that treat visual sequences as unified logical structures rather than unordered collections of frames.

Related papers

• Interpreting Object-Dependent Concept Brittleness in Text-to-Image Diffusion Models [105.53190946111586]
  We term this phenomenon object-dependent concept brittleness.<n>We present an interpretability-oriented framework to audit and minimally correct these failures.<n>We evaluate the proposed framework on style and attribute failure cases across multiple diffusion backbones.
  arXiv  Detail & Related papers  (2026-09-09T09:07:40Z)
• What's the Catch? Evaluating Temporal Consistency in Vision-Language Models [0.11121828451787184]
  Vision-language models (VLMs) achieve strong performance on video and image-sequence benchmarks, yet it remains unclear whether they capture temporal structure.<n>We introduce TimeCatch, where temporal anomalies are created by swapping consecutive frames and frame-level anomalies by replacing a frame with Gaussian noise.<n>Our evaluation reveals a substantial gap between frame-level and temporal anomaly detection.
  arXiv  Detail & Related papers  (2026-08-24T16:40:49Z)
• Temporally-Aligned Evaluation for Audio-Driven Talking Head Generation [28.732587811107777]
  Existing evaluation protocols mainly rely on frame-wise metrics that assume strict temporal correspondence between generated and reference videos.<n>We introduce a unified sequence-level reformulation that integrates Soft Dynamic Time Warping into established evaluation pipelines.<n>We show that frame-wise evaluation can be viewed as a special case under rigid alignment, while sequence-level alignment provides improved stability, lower sensitivity to timing differences, and clearer separation between modeling paradigms.
  arXiv  Detail & Related papers  (2026-05-31T05:44:42Z)
• Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models [154.39583176906893]
  Large language models (MLs) have advanced understanding, they remain highly prone to hallucinations in scenes.<n>We argue this stems from a failure intemporal monitoring, the ability to persistently track object dynamic identities, states, and relations over time.<n>Existing benchmarks obscure this deficit relying on single final-answer queries that can often be resolved via local visual cues or statistical priors.
  arXiv  Detail & Related papers  (2026-05-09T14:32:36Z)
• The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation [18.83395586542378]
  Multimodal Large Language Models (MLLMs) are increasingly integrated with Retrieval-Augmented Generation (RAG)<n>We identify and formalize the phenomenon of recorruption, where the introduction of even perfectly accurate "oracle" context causes a capable model to abandon an initially correct prediction.<n>We propose Bottleneck Attention Intervention for Recovery (BAIR), a parameter-free, inference-time framework that restores visual saliency and applies position-aware penalties to textual distractors.
  arXiv  Detail & Related papers  (2026-05-07T02:27:04Z)
• A Progressive Training Strategy for Vision-Language Models to Counteract Spatio-Temporal Hallucinations in Embodied Reasoning [49.61652671596548]
  "Multi-image hallucination reasoning", where a massive performance drop between forward and temporal queries reveals a dependence on superficial shortcuts instead of genuine understanding.<n>To mitigate this, we develop a new dataset: Chain-of-temporal building on detailed reasoning into Chain steps and definitive judgments.<n>Our experiments demonstrate that this approach not only improves accuracy but also the forward-backward performance gap from over 70% to only 6.53%.
  arXiv  Detail & Related papers  (2026-04-12T07:48:44Z)
• Anatomy of a Lie: A Multi-Stage Diagnostic Framework for Tracing Hallucinations in Vision-Language Models [62.932580559941414]
  Vision-Language Models (VLMs) frequently "hallucinate" - generate plausible yet factually incorrect statements.<n>We propose a new paradigm for diagnosing hallucinations, recasting them from static output errors into dynamic pathologies of a model's computational cognition.
  arXiv  Detail & Related papers  (2026-03-16T17:20:38Z)
• NarraScore: Bridging Visual Narrative and Musical Dynamics via Hierarchical Affective Control [59.6128550986024]
  NarraScore is a hierarchical framework predicated on the core insight that emotion serves as a high-density compression of narrative logic.<n>NarraScore employs a Dual-Branch Injection strategy to reconcile global structure with local dynamism.<n>NarraScore achieves state-of-the-art consistency and narrative alignment with negligible computational overhead.
  arXiv  Detail & Related papers  (2026-02-09T09:39:42Z)
• Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts [74.47786985522762]
  We identify a critical failure mode termed textual inertia, where models tend to blindly adhere to the erroneous text while neglecting conflicting visual evidence.<n>We propose the LogicGraph Perturbation Protocol that structurally injects perturbations into the reasoning chains of diverse LMMs.<n>Results reveal that models successfully self-correct in less than 10% of cases and predominantly succumb to blind textual error propagation.
  arXiv  Detail & Related papers  (2026-01-07T16:39:34Z)
• NarrativeTrack: Evaluating Video Language Models Beyond the Frame [10.244330591706744]
  We introduce NarrativeTrack, the first benchmark to evaluate narrative understanding in MLLMs.<n>We decompose videos into constituent entities and examine their continuity via a Compositional Reasoning (CRP) framework.<n>CRP challenges models to advance from temporal persistence to contextual evolution and fine-grained perceptual reasoning.
  arXiv  Detail & Related papers  (2026-01-03T07:12:55Z)
• Causality Matters: How Temporal Information Emerges in Video Language Models [17.570777893613137]
  We find that removing or modifying positional encodings in video inputs yields minimal degradation in the performance of temporal understanding.<n>To explain this behavior, we conduct substantial analysis experiments to trace how temporal information is integrated within the model.<n>We propose two efficiency-oriented strategies: staged cross-modal attention and a temporal exit mechanism for early token truncation.
  arXiv  Detail & Related papers  (2025-08-15T16:33:14Z)

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.