paper-with-me

Papers

ATHENA: Accelerated Multi-Task Heterogeneous Influence Functions for Robot Data Curation

2026-06-15 · Tao Xu, Jiaxin Wang, Runhao Zhang, Jiayi Guan, Xianchao Zeng, Weixi Song, Xinyu Zhou, Zhetao Chen, Guang Chen, Yong-Lu Li arxiv

In robot imitation learning, influence functions provide a principled approach to quantify each demonstration's effect on robot task outcomes, yet scaling them to billion-parameter Vision-Language-Action (VLA) models is limited by computational and multitask bottlenecks. To this end, we propose ATHENA, an influence function framework tailored for multitask VLA data curation at a billion-parameter scale. Concretely, it leverages the Kronecker structure of linear-layer gradients to reduce projection cost, and approximates dense Hessian inversion with a rank-r Random Truncated Approximation, achieving about a 313.4x speedup in influence computation. Furthermore, ATHENA formulates global and local interactive influence to balance data curation across 50 jointly trained tasks. Extensive evaluations on RoboTwin 2.0 and real-robot deployment, covering 9.34 and 6.90 hours of demonstrations, respectively, show that ATHENA matches or exceeds full-data joint fine-tuning using only 50% of demonstrations in simulation and 66.7% of data across six real-robot tasks. Overall, ATHENA demonstrates its effectiveness for data curation in billion-parameter multitask VLA fine-tuning.

📄 PDF Abstract BibTeX arXiv:2606.16208

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

When is ACL's Deadline? A Scientific Conversational Agent

2019-11-23 · Mohsen Mesgar, Paul Youssef, Lin Li, Dominik Bierwirth 외

Our conversational agent UKP-ATHENA assists NLP researchers in finding and exploring scientific literature, identifying relevant authors, planning or post-processing conference visits, and preparing paper submissions usi…

Personalized Decision Modeling: Utility Optimization or Textualized-Symbolic Reasoning

2025-11-04 · Yibo Zhao, Yang Zhao, Hongru Du, Hao Frank Yang arxiv

Decision-making models for individuals, particularly in high-stakes scenarios like vaccine uptake, often diverge from population optimal predictions. This gap arises from the uniqueness of the individual decision-making …

ATHENA: Agentic Team for Hierarchical Evolutionary Numerical Algorithms

2025-12-03 · Juan Diego Toscano, Daniel T. Chen, George Em Karniadakis arxiv

Progress in computational science depends on complex numerical workflows that must faithfully encode physical laws, yet translating conceptual insight into reliable code remains a major bottleneck. Although large languag…

A Transformer-based Response Evaluator for Open-Domain Spoken Conversation

2023-02-09 · Vrindavan Harrison, Rishi Rajasekaran, Marilyn Walker

Many open-domain dialogue systems rely on multiple response generators, any of which can contribute a response to the dialogue in a particular context. Thus the ability to compare potential responses and then select the …

Athena: Synergizing Data Prefetching and Off-Chip Prediction via Online Reinforcement Learning

2026-01-24 · Rahul Bera, Zhenrong Lang, Caroline Hengartner, Konstantinos Kanellopoulos 외 arxiv

Prefetching and off-chip prediction are two techniques proposed to hide long memory access latencies in high-performance processors. In this work, we demonstrate that: (1) prefetching and off-chip prediction often provid…

Reinforcement Learning