arxiv arXiv cs.AI · 51d ago · research

Finetuning VLA Models Requires Fewer Layers Than Thought

from English

Vision-Language-Action models show severe layer-wise redundancy despite large parameter counts. A training-free compression method using Centered Kernel Alignment removes twin layers, reducing model depth by up to 50% and enabling 40-50% faster training and up to 30% faster inference without performance loss, validated across simulation and real-world robotic tasks.

Importance 3/3 New feature vs. leaders New harness with differentiators arXiv cs.AI Google DeepMind Microsoft Research NVIDIA AI agents Reasoning models Robotics

Read original

arxiv arXiv cs.LG · 51d ago · 1 view

Pose6DAug: Physically Plausible Multi-view Object Swapping

Pose6DAug enables robot data augmentation by swapping objects in successful episodes while preserving physically valid 6D pose trajectories. It operates in 3D using a mesh anchored by temporally coherent poses, ensuring multi-view consistency and physical plausibility. Fine-tuning a VLA policy on this augmented data improves novel object success rates by 16.5% over state-of-the-art baselines.

arxiv arXiv cs.LG · 54d ago · 2 views

ROVE: Reinforcement Learning with Human Interventions for Humanoid Manipulation

ROVE enables humanoid Vision-Language-Action models to learn effective manipulation behaviors using imperfect human interventions. It combines a human-in-the-loop data collection pipeline with Optimistic Value Estimation and cross-embodiment supervision to prioritize high-value actions and improve robustness. ROVE outperforms baseline methods on real-world, contact-rich manipulation tasks through iterative rollout and intervention cycles.

arxiv arXiv cs.LG · 51d ago · 2 views

Probe-and-Refine Tuning Improves Coding Agent Performance

A new method called probe-and-refine tuning uses synthetic bug-fix probes to iteratively improve repository guidance files with single-shot LLM calls, without agent loops or tool use. On SWE-bench Verified, it achieves a 33.0% mean resolve rate—14.5 percentage points higher than the initial static knowledge base—showing improved coverage rather than patch precision. The method enables agents to use larger step budgets effectively, and performance remains stable across models when diagnostic output is sufficient.

arxiv arXiv cs.LG · 51d ago · 3 views

Execution-State Capsules for Low-Latency On-Device AI Serving

Execution-state capsules enable graph-bound checkpointing and restoration of complete execution state, including KV, recurrent, and convolution states, for low-latency, small-batch on-device AI serving. On RTX 5090 and Jetson AGX Thor, capsule restore achieves byte-exact and token-identical correctness, with sub-millisecond GPU operations and TTFT speedups up to 27x at 16k tokens, demonstrating significant latency reduction in interactive AI workflows.

arxiv arXiv cs.AI · 51d ago

Lagrange: Open-Vocabulary Sparse Framework for End-to-End Driving

Lagrange introduces an open-vocabulary, energy-based sparse framework for generalized end-to-end driving. It uses Vision-Language Models to generate class-agnostic object proposals and encodes them into continuous semantic tokens, enabling robust generalization to anomalous scenarios while adhering to vehicle kinematics through Lagrangian action minimization.