← Back to all news
arXiv AI By Mingxuan Yan, Litian Gong, Dihong Huang, Zhixu Li, Tianyu Zhang, Wenqian Zhang, Hefeifei Jiang, Peihao Li, Jian Zhang, Mengfei Zhao, Yikai Tang, Hai Zhai, Zehao Wang, Ruijian Liang, Yanjia Huang, Lin Shao, Changliu Liu, Kaiyu Hang, Zhiwen Fan, Masayoshi Tomizuka, Jianfei Yang, Jiachen Li

What Matters in Quality-Aware VLA Pretraining: A Benchmark and Empirical Study

Read the original on arXiv AI →

The paper investigates how to best utilize demonstration quality during vision‑language‑action (VLA) pretraining to boost downstream task performance. It compares weighting, filtering, and conditioning strategies across different reward definitions and label granularities, using 5,000 suboptimal human demonstrations and 24 target tasks while keeping demonstration sets and training budgets matched. The study introduces AXIS‑Bench, a benchmark built on crowdsourced simulation demonstrations that allows recomputation of quality labels and replayable trajectories for both simulation and real‑world tasks, showing that simulation pretraining can raise real‑world success rates from 9.6% to 31.7%.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Jun 16

Retrieve, Don't Retrain: Extending Vision Language Action Models to New Tasks at Test Time

arXiv:2606. 15631v1 Announce Type: cross Abstract: Extending a vision-language-action (VLA) policy to a new task typically requires task-specific teleoperated demonstrations and per-task fine-tuning, making adaptation costly in both data collection and compute.

By Jeongeun Park, Juhan Park, Taekyung Kim, Sungjoon Choi, Dongyoon Han, Sangdoo Yun
ragroboticsfine-tuningmultimodal
More like this →
arXiv Computer Vision
Oct 12

TraceICL: Thread-Resolved Demonstration Curation with Consistency Calibration for Robust Multimodal In-Context Learning

The paper introduces AVA, a training‑free framework that improves interleaved multimodal in‑context learning for frozen large vision‑language models. AVA decomposes the interleaved context into per‑sample threads, selects influential demonstrations while removing redundancy, and calibrates consistency using a weak sibling model’s confidence. Across five vision‑language benchmarks, AVA raises performance by up to 2.4 points, narrows the visual performance gap on ScienceQA‑IMG, and remains robust even with 30% noisy demonstrations, all while keeping the LVLM frozen and adding only modest latency.

By Yigit Yilmaz, Elena Petrova, Mehmet Kaya, Lucia Rossi, Amir Rahman
llmsmultimodalbenchmarks
More like this →
arXiv Computer Vision
1d ago

When Vision Overrides Language: Evaluating and Mitigating Counterfactual Failures in VLAs

Vision‑Language‑Action models (VLAs) often fail to follow language instructions because they rely on vision shortcuts learned from biased datasets. The authors introduce LIBERO‑CF, a counterfactual benchmark that tests language following by providing alternative task instructions for the same scene layouts, revealing widespread counterfactual failures. They propose Counterfactual Action Guidance (CAG), a dual‑branch inference scheme that amplifies language conditioning, which consistently improves language grounding and task success across diverse VLAs without requiring extra data or architectural changes.

By Yu Fang, Yuchun Feng, Dong Jing, Jiaqi Liu, Yue Yang, Zhenyu Wei, Daniel Szafir, Mingyu Ding
roboticsmultimodalbenchmarks
More like this →
arXiv AI
Jun 2

VLM4VLA: Revisiting Vision-Language-Models in Vision-Language-Action Models

arXiv:2601. 03309v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models, which integrate pretrained large Vision-Language Models (VLM) into their policy backbone, are gaining significant attention for their promising generalization capabilities.

By Jianke Zhang, Xiaoyu Chen, Qiuyue Wang, Mingsheng Li, Yanjiang Guo, Yucheng Hu, Jiajun Zhang, Shuai Bai, Junyang Lin, Jianyu Chen
llmscomputer-visionroboticsfine-tuningmultimodalbenchmarks
More like this →