4arXiv cs.AI (Artificial Intelligence)·
VAORA: Visual Action Outcome Reasoning Alignment for physical reasoning generalization in VLMs
Researchers introduce VAORA (Visual Action Outcome Reasoning Alignment), a reward design framework targeting two failure modes in vision-language models applied to interactive physical reasoning: hallucinated chain-of-thought reasoning and misalignment between reasoning and actions. VAORA uses two complementary rewards — a Visual Alignment Reward anchoring reasoning to visual context, and a Visual-Action Alignment Reward grounding reasoning in action outcomes — plus dense rewards from a pre-trained expert agent for training stability. Experiments on the PHYRE and Virtual Tool benchmarks show improved generalization to novel tasks and unseen environments.
Related guides (2)
Related events (8)
CORA: Consistency-Oriented Reasoning Alignment addresses thinking-answer gap in multimodal RLVR
Researchers identify and analyze a systematic inconsistency between reasoning traces and final answers in RLVR-trained large vision-language models, showing the problem persists throughout GRPO training and inference. They propose CORA, which introduces a lightweight plug-and-play consistency reward model and a Hybrid Reward Advantage Splitting (HRAS) mechanism to coordinate task and consistency optimization. Experiments across multimodal reasoning benchmarks show CORA improves both task performance and reasoning faithfulness.
VRRL: Reinforcement learning framework for visually grounded self-reflection in vision-language models
Researchers propose VRRL, a reinforcement learning training framework designed to improve self-reflection in large vision-language models (LVLMs) by ensuring corrections are grounded in visual inputs rather than purely textual reasoning. The framework uses two mechanisms: random masking of trajectory prefixes to emphasize error recovery, and buffered roll-ins from an experience replay buffer to expose models to diverse failure states. Evaluated on visual grounding tasks (tables, charts) and spatial navigation benchmarks, VRRL substantially improves out-of-distribution accuracy over standard RL and reflection-oriented fine-tuning baselines. The work addresses a concrete failure mode where LVLMs lose visual grounding during chain-of-thought reflection.
Act2Answer: Benchmarking commonsense and world knowledge retention in Vision-Language-Action models
Researchers introduce Act2Answer, a protocol for evaluating how much commonsense and factual knowledge VLA models retain after fine-tuning on robotics data. The approach converts knowledge benchmark questions into tabletop object-placement episodes, yielding action-grounded success rates that reduce confounds from low-level control failures. A large-scale study of 7 VLA models and 9 VLM baselines finds that VLAs retain solid performance on simple concepts but show larger gaps on richer semantic categories compared to their source VLMs, and that VQA co-training is associated with better knowledge retention.
Vision Language Model Alignment in TRL
Hugging Face's TRL library has added support for aligning Vision Language Models (VLMs), extending existing RLHF and preference optimization tooling to multimodal settings. The blog post covers the new capabilities for training VLMs with alignment techniques such as DPO and related methods. This expands the open-source ecosystem for multimodal model fine-tuning and alignment.
Two-stage action prior pretraining improves cross-embodiment VLA robot manipulation
Researchers propose a two-stage training framework for Vision-Language-Action (VLA) models that pretrains the action module with motion priors before cross-modal alignment begins. Stage 1 uses a flow-matching-based encoder-decoder to learn temporal motion structure from unconditioned action trajectories alone; Stage 2 transfers this prior to VLA training via decoder reuse and latent distillation. Evaluated across 13 cross-embodiment tasks in simulation and real-world settings, the approach achieves faster convergence, higher success rates, and notably better performance in data-scarce real-world scenarios compared to VLA training without action priors.
VEPO: Vision-anchored token selection improves RL for visual reasoning
A new arXiv paper identifies a failure mode of entropy-based credit assignment in multimodal reinforcement learning: vision-sensitive tokens with naturally low entropy are systematically ignored, causing the mechanism to collapse in visual reasoning tasks. The authors propose VEPO (Vision-Entropy token-selection for Policy Optimization), which couples visual sensitivity with token entropy via a multiplicative scheme to redirect gradient credit toward tokens that are both visually grounded and semantically informative. VEPO outperforms entropy-only baselines by 2.28 points at 7B scale and 3.15 points at 3B scale on visual reasoning benchmarks.
LabVLA: Vision-Language-Action model and RoboGenesis data engine for scientific laboratory robotics
Researchers introduce LabVLA, a Vision-Language-Action model designed to bridge written scientific protocols and physical robot execution in laboratory settings. To address the data scarcity problem, they build RoboGenesis, a simulation-based data engine that composes lab workflows from atomic skills and generates structured demonstrations across robot embodiments. LabVLA uses a two-stage training recipe combining FAST action token pretraining on a Qwen3-VL-4B-Instruct backbone with flow matching posttraining via a DiT action expert. On the LabUtopia benchmark, LabVLA achieves the highest average success rate among evaluated baselines in both in-distribution and out-of-distribution settings.
VADAOrchestra: Neurosymbolic framework combining LLM orchestration with Datalog+/- symbolic reasoning for adaptive workflows
VADAOrchestra is a neurosymbolic framework that hybridizes LLM-based orchestration with a symbolic Datalog+/- inference engine to model complex, adaptive workflows. An LLM incrementally plans and encodes workflow steps as logic programs, while a dedicated symbolic engine handles all inference, decoupling orchestration from execution. The approach targets auditability, scalability over large datasets, and explainability — limitations of pure LLM-agent architectures — and is evaluated on real-world financial use cases. The work positions itself as a bridge between traditional Business Process Management rigidity and LLM agent flexibility.