Cross-Stage Coherence in Hierarchical Driving VQA: Explicit Baselines and Learned Gated Context Projectors

• URL: http://arxiv.org/abs/2604.22560v1
• Date: Fri, 24 Apr 2026 13:54:48 GMT
• Title: Cross-Stage Coherence in Hierarchical Driving VQA: Explicit Baselines and Learned Gated Context Projectors
• Abstract summary: We present a comparative study of cross-stage context passing on DriveLM-nuScenes using two complementary mechanisms.<n>The explicit variant evaluates three prompt-based conditioning strategies on a domain-adapted 4B VLM without additional training.<n>The implicit variant introduces gated context projectors, which extract a hidden-state vector from one stage and inject a normalized, gated projection into the next stage's input embeddings.
• Score: 0.688204255655161
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Graph Visual Question Answering (GVQA) for autonomous driving organizes reasoning into ordered stages, namely Perception, Prediction, and Planning, where planning decisions should remain consistent with the model's own perception. We present a comparative study of cross-stage context passing on DriveLM-nuScenes using two complementary mechanisms. The explicit variant evaluates three prompt-based conditioning strategies on a domain-adapted 4B VLM (Mini-InternVL2-4B-DA-DriveLM) without additional training, reducing NLI contradiction by up to 42.6% and establishing a strong zero-training baseline. The implicit variant introduces gated context projectors, which extract a hidden-state vector from one stage and inject a normalized, gated projection into the next stage's input embeddings. These projectors are jointly trained with stage-specific QLoRA adapters on a general-purpose 8B VLM (InternVL3-8B-Instruct) while updating only approximately 0.5% of parameters. The implicit variant achieves a statistically significant 34% reduction in planning-stage NLI contradiction (bootstrap 95% CIs, p < 0.05) and increases cross-stage entailment by 50%, evaluated with a multilingual NLI classifier to account for mixed-language outputs. Planning language quality also improves (CIDEr +30.3%), but lexical overlap and structural consistency degrade due to the absence of driving-domain pretraining. Since the two variants use different base models, we present them as complementary case studies: explicit context passing provides a strong training-free baseline for surface consistency, while implicit gated projection delivers significant planning-stage semantic gains, suggesting domain adaptation as a plausible next ingredient for full-spectrum improvement.

Related papers

• SeGDeP: Semantic- and Geometric-Aware Decoupled Prompts for Reasoning Segmentation [51.46828526392219]
  Reasoning segmentation converts an implicit linguistic conclusion into a precise mask.<n>Existing MLLM-segmenter interfaces either use a special trigger or compress both signals into one context.<n>We present SeGDeP, an explicit what-where interface.
  arXiv  Detail & Related papers  (2026-09-08T15:11:32Z)
• A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM [40.07171947823021]
  Chain-of-Thought (CoT) improves the reasoning ability of Large Language Models (LLMs) but incurs substantial computation and context costs.<n>We present A*-Thought-V2, a geometric dynamics of LLM guided framework that models CoT as a hidden-state trajectory and replaces hard deletion with an explicit-implicit interleaved latent architecture.
  arXiv  Detail & Related papers  (2026-09-07T17:56:20Z)
• Architecture-Dependent Causal Transfer of Activation States Across Large Language Models [51.56484100374058]
  Internal language activation states can be transferred causally between different large language models.<n>Injecting projected activations into a target model during generation produces a statistically significant, preregistered causal effect.<n>We conclude that end-to-end activation-state transfer between LLMs, as currently implemented, is architecture-dependent rather than universal.
  arXiv  Detail & Related papers  (2026-08-17T09:53:27Z)
• SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning [99.49739831030388]
  We study how the design of this interface shapes the agent's capacity for open-ended spatial reasoning.<n>We propose SpatialClaw, a training-free framework for spatial reasoning that adopts code as the action interface.
  arXiv  Detail & Related papers  (2026-06-11T17:59:36Z)
• Routing-Aligned Fine-Tuning for Multilingual Downstream Tasks in Mixture-of-Experts Models [22.38900139129832]
  Existing fine-tuning approaches treat MoE models as monolithic learners, ignoring the heterogeneous routing structure that develops during pretraining.<n>We propose RA-MoE (Routing-Aligned MoE Fine-Tuning), a three-stage framework that categorizes parallel task examples into a four-way taxonomy based on correctness in English and the target language.<n>Experiments across three MoE models, three tasks, and six target languages demonstrate that RA-MoE consistently outperforms standard SFT and strong baselines.
  arXiv  Detail & Related papers  (2026-05-27T11:01:25Z)
• PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement [22.285969725581044]
  Large language model (LLM)-based agents generate seemingly coherent plans that fail upon execution due to infeasible actions, constraint violations, and compounding errors over extended horizons.<n>PIVOT addresses this plan-execution misalignment through a self-supervised framework that treats trajectories as optimizable objects iteratively refined via environment interaction.
  arXiv  Detail & Related papers  (2026-05-11T20:43:20Z)
• NCL-BU at SemEval-2026 Task 3: Fine-tuning XLM-RoBERTa for Multilingual Dimensional Sentiment Regression [5.8096845806519335]
  This paper describes a system developed for Track A - Subtask 1 (Dimensional Aspect Sentiment Regression)<n>It aims to predict real-valued VA scores in the [1, 9] range for each given aspect in a text.
  arXiv  Detail & Related papers  (2026-04-10T03:38:53Z)
• VLS: Steering Pretrained Robot Policies via Vision-Language Models [31.189909515514668]
  Vision-Language Steering (VLS) is a training-free framework for inference-time adaptation of frozen generative robot policies.<n>VLS treats adaptation as an inference-time control problem, steering the sampling process of a pretrained diffusion or flow-matching policy.
  arXiv  Detail & Related papers  (2026-02-03T19:50:16Z)
• Model Specific Task Similarity for Vision Language Model Selection via Layer Conductance [92.72779885657373]
  We propose a framework that grounds model selection in the internal functional dynamics of the visual encoder.<n>Our approach represents each task via layer wise conductance and derives a target-conditioned block importance distribution through entropy regularized alignment.<n>Building on this, we introduce Directional Conductance Divergence (DCD), an asymmetric metric that quantifies how effectively a source task covers the target's salient functional blocks.
  arXiv  Detail & Related papers  (2026-02-01T17:29:43Z)
• Subspace Alignment for Vision-Language Model Test-time Adaptation [82.83192844597593]
  Vision-language models (VLMs) are vulnerable to distribution shifts.<n>Existing test-time adaptation methods rely on zero-shot predictions as pseudo-labels for self-training.<n>We propose SubTTA, which aligns the semantic subspaces of both modalities to enhance zero-shot predictions.
  arXiv  Detail & Related papers  (2026-01-13T02:02:41Z)
• Prune-Then-Plan: Step-Level Calibration for Stable Frontier Exploration in Embodied Question Answering [52.69447404069251]
  Large vision-language models (VLMs) have improved embodied question answering (EQA) agents by providing strong semantic priors for open-vocabulary reasoning.<n>We propose Prune-Then-Plan, a framework that stabilizes exploration through step-level calibration.
  arXiv  Detail & Related papers  (2025-11-24T22:50:50Z)
• Enhancing the Outcome Reward-based RL Training of MLLMs with Self-Consistency Sampling [90.87033586963828]
  Outcome-reward reinforcement learning (RL) is a common and increasingly significant way to refine the step-by-step reasoning of multimodal large language models (MLLMs)<n>We propose Self-Consistency Sampling (SCS) to correct this issue.<n>Based on Qwen2.5-VL-7B-Instruct, SCS improves accuracy by up to 7.7 percentage points on six multimodal benchmarks with negligible extra computation.
  arXiv  Detail & Related papers  (2025-11-13T18:59:57Z)
• Data Trajectory Alignment for LLM Domain Adaptation: A Two-Phase Synthesis Framework for Telecommunications Mathematics [6.653834890554154]
  General-purpose large language models (LLMs) are increasingly deployed in verticals such as telecommunications.<n>We propose Data Trajectory Alignment (DTA), a two-phase, model-agnostic data curation framework.<n>DTA treats solution processes - not only final answers - as first-class supervision.
  arXiv  Detail & Related papers  (2025-11-10T07:05:08Z)
• Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail [85.47497935739936]
  Alpamayo-R1 (AR1) is a vision-language-action model that integrates Chain of Causation reasoning with trajectory planning.<n>We show AR1 achieves 12% improvement in planning accuracy on challenging cases compared to a trajectory-only baseline.<n>We plan to release AR1 models and a subset of the CoC in a future update.
  arXiv  Detail & Related papers  (2025-10-30T01:25:34Z)
• CLGRPO: Reasoning Ability Enhancement for Small VLMs [4.551310348498266]
  Small Vision Language Models (SVLMs) generally refer to models with parameter sizes less than or equal to 2B.<n>This paper proposes a post-training optimization paradigm called the Incremental Training Strategy to enhance the reasoning ability of SVLMs.<n> Experimental results demonstrate that our method significantly improves the reasoning ability of 1B SVLM.
  arXiv  Detail & Related papers  (2025-06-22T14:32:15Z)
• Prismatic VLMs: Investigating the Design Space of Visually-Conditioned Language Models [73.40350756742231]
  Visually-conditioned language models (VLMs) have seen growing adoption in applications such as visual dialogue, scene understanding, and robotic task planning. Despite the volume of new releases, key design decisions around image preprocessing, architecture, and optimization are under-explored.
  arXiv  Detail & Related papers  (2024-02-12T18:21:14Z)

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.