VLMs Trace Without Tracking: Diagnosing Failures in Visual Path Following

• URL: http://arxiv.org/abs/2605.15672v1
• Date: Fri, 15 May 2026 06:48:53 GMT
• Title: VLMs Trace Without Tracking: Diagnosing Failures in Visual Path Following
• Abstract summary: Vision-language models (VLMs) achieve strong performance on multimodal benchmarks, but may still lack robust control over basic visual operations.<n>We study textitline tracing, where a model must follow a selected visual path through successive local continuations.<n>We design controlled tracing tasks that introduce nearby competitors while reducing semantic and topological ambiguity.
• Score: 10.22092497929711
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-language models (VLMs) achieve strong performance on multimodal benchmarks, but may still lack robust control over basic visual operations. We study \textit{line tracing}, where a model must follow a selected visual path through successive local continuations. To isolate this ability, we design controlled tracing tasks that introduce nearby competitors while reducing semantic and topological ambiguity such as crossings and overlaps. Across these tasks, even state-of-the-art VLMs frequently lose the target path and switch to nearby alternatives, especially when those alternatives look locally similar to the target. Behavioral interventions and internal analyses indicate that these failures arise from local competition: nearby similar distractors pull the model away from the true continuation. Standard remedies do not remove this bottleneck: model-size scaling provides only limited gains, reasoning partially compensates through costly substitute strategies, and explicit tracing instructions fail to recover stable path following. Finally, tests on tangled-cable scenes and metro maps with richer visual complexity show that the same path-switching failure persists beyond our controlled settings.

Related papers

• AutoPath: Learning Transferable Goal-Conditioned Stochastic Path Prior for Safe Navigation Without Human Demonstrations [103.61864552260316]
  Real-time navigation in dynamic environments requires collision-free and dynamically feasible motion under limited perception.<n>We formulate navigation as learning a transferable goal-aligned path prior that models a reusable distribution over goal-aligned geometry-consistent local paths conditioned on local observations.<n>We develop a structured prior learning framework that parameterizes local paths using a geometry-aware polar action manifold and incorporates risk-sensitive utility shaping with multi-goal distributional rollouts for stable and safety-aware planning.
  arXiv  Detail & Related papers  (2026-07-13T16:00:41Z)
• Focus When Necessary: Adaptive Routing and Collaborative Grounding for Training-Free Visual Grounding [56.38477453373368]
  LazyMCoT is a dynamic and training-free framework that adaptively allocates visual grounding efforts based on sample difficulty.<n>LazyMCoT rivals training-based approaches by simultaneously improving reasoning accuracy and reducing average inference latency.
  arXiv  Detail & Related papers  (2026-06-15T03:17:44Z)
• Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring [35.34120853605602]
  textbf-and-Seek is a framework that formulates VLA failure detection as a coarsely supervised learning problem.<n>We evaluate Hide-and-Seek on LIBERO, VLABench, and a real-world robotic platform across three representative VLA policies.
  arXiv  Detail & Related papers  (2026-05-29T04:40:12Z)
• TraversalBench: Challenging Paths to Follow for Vision Language Models [11.032462608031922]
  Vision-language models (VLMs) perform strongly on multimodal benchmarks, but the ability to follow complex visual paths is under-tested.<n>We introduce TraversalBench, a benchmark for exact visual path traversal.<n>We find that self-intersections are the dominant source of difficulty.
  arXiv  Detail & Related papers  (2026-04-13T04:58:52Z)
• TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models [59.13964209628383]
  Vision-Language--Action (VLA) policies have shown strong progress in mapping language instructions and visual observations to robotic actions, yet their reliability degrades in cluttered scenes with distractors.<n>We propose TAG (Target-Agnostic Guidance), a simple inference-time guidance mechanism that explicitly reduces distractor- and appearance-induced bias in VLA policies.<n>We evaluate TAG on standard manipulation benchmarks, including LIBERO, LIBERO-Plus, and VLABench, where it consistently improves robustness under clutter and reduces near-miss and wrong-object executions.
  arXiv  Detail & Related papers  (2026-03-25T17:56:32Z)
• HaltNav: Reactive Visual Halting over Lightweight Topological Priors for Robust Vision-Language Navigation [1.774434289475737]
  Vision-and-Language Navigation (VLN) is shifting from rigid, step-by-step instruction following toward open-vocabulary, goal-oriented autonomy.<n>We propose a hierarchical navigation framework HaltNav that couples the robust global planning of osmAG with the local exploration and instruction-grounding capability of VLN.
  arXiv  Detail & Related papers  (2026-03-13T06:22:35Z)
• Let's Reward Step-by-Step: Step-Aware Contrastive Alignment for Vision-Language Navigation in Continuous Environments [38.97818584066075]
  Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to learn complex reasoning from long-horizon human interactions.<n>Current training paradigms struggle to balance generalization capability, error recovery and training stability.<n>We introduce Step-Aware Contrastive Alignment (SACA), a framework designed to extract dense supervision from imperfect trajectories.
  arXiv  Detail & Related papers  (2026-03-10T14:45:50Z)
• TopoCurate:Modeling Interaction Topology for Tool-Use Agent Training [53.93696896939915]
  Training tool-use agents typically rely on Supervised Fine-Tuning (SFT) on successful trajectories and Reinforcement Learning (RL) on pass-rate-selected tasks.<n>We propose TopoCurate, an interaction-aware framework that projects multi-trial rollouts from the same task into a unified semantic quotient topology.<n>TopoCurate achieves consistent gains of 4.2% (SFT) and 6.9% (RL) over state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-03-02T10:38:54Z)
• ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and Localization [62.03035862528452]
  ForgeryVCR is a framework that materializes imperceptible traces into explicit visual intermediates via Visual-Centric Reasoning.<n>ForgeryVCR achieves state-of-the-art (SOTA) performance in both detection and localization tasks.
  arXiv  Detail & Related papers  (2026-02-15T11:14:47Z)
• Contamination Detection for VLMs using Multi-Modal Semantic Perturbation [73.76465227729818]
  Open-source Vision-Language Models (VLMs) have achieved state-of-the-art performance on benchmark tasks.<n>Pretraining corpora raise a critical concern for both practitioners and users: inflated performance due to test-set leakage.<n>We show that existing detection approaches either fail outright or exhibit inconsistent behavior.<n>We propose a novel simple yet effective detection method based on multi-modal semantic perturbation.
  arXiv  Detail & Related papers  (2025-11-05T18:59:52Z)
• Generative Point Tracking with Flow Matching [32.15342097497571]
  We introduce Generative Point Tracker (GenPT), a generative framework for modelling multi-modal trajectories.<n>GenPT is trained with a novel flow matching formulation that combines the iterative refinement of discriminative trackers.<n>We show how our model's generative capabilities can be leveraged to improve point trajectory estimates.
  arXiv  Detail & Related papers  (2025-10-23T19:25:14Z)
• Towards Unified Token Learning for Vision-Language Tracking [65.96561538356315]
  We present a vision-language (VL) tracking pipeline, termed textbfMMTrack, which casts VL tracking as a token generation task. Our proposed framework serializes language description and bounding box into a sequence of discrete tokens. In this new design paradigm, all token queries are required to perceive the desired target and directly predict spatial coordinates of the target.
  arXiv  Detail & Related papers  (2023-08-27T13:17:34Z)
• Zero-Shot Temporal Action Detection via Vision-Language Prompting [134.26292288193298]
  We propose a novel zero-Shot Temporal Action detection model via Vision-LanguagE prompting (STALE) Our model significantly outperforms state-of-the-art alternatives. Our model also yields superior results on supervised TAD over recent strong competitors.
  arXiv  Detail & Related papers  (2022-07-17T13:59:46Z)

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.