Papers arxiv:2606.05165 STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations Published on Jun 3 Submitted by Rishit Dagli on Jun 4 Authors: Rishit Dagli , , , , , , Abstract STRIDE framework enables efficient training data attribution for LLMs by modeling functional effec… …
… Motivated by this gap, we propose GUI-RobustEval, a fine-grained benchmark for measuring error awareness and recovery across 11 error types and 4 error depths, and RoTS, a tree-based online synthesis framework that generates 800K error-recovery trajectories by actively exploring failure modes and s… …
Papers arxiv:2604.25441 Praxy Voice: Voice-Prompt Recovery + BUPS for Commercial-Class Indic TTS from a Frozen Non-Indic Base at Zero Commercial-Training-Data Cost Published on Apr 28 Submitted by Venkata Pushpak Teja Menta on Apr 30 Praxel Authors: Venkata Pushpak Teja Menta Abstract Researchers e… …
… No dataset linking this paper Cite arxiv.org/abs/2606.02398 in a dataset README.md to link it from this page. …
… Finding: it's the trusted checkpoint state that does the recovery work, not any specific prefix tail. 🤗 Dataset: https://huggingface.co/datasets/parsa-mz/puzzlezoo 💻 Code: https://github.com/parsa-mz/RePot This is an automated message from the Librarian Bot . …
… No dataset linking this paper Cite arxiv.org/abs/2605.24047 in a dataset README.md to link it from this page. …
… Get this paper in your agent: hf papers read 2605.09530 Don't have the latest CLI? curl -LsSf https://hf.co/cli/install.sh | bash No dataset linking this paper Cite arxiv.org/abs/2605.09530 in a dataset README.md to link it from this page. …
… No dataset linking this paper Cite arxiv.org/abs/2605.26029 in a dataset README.md to link it from this page. …
… View arXiv page View PDF Add to collection Community ENERGYFLOW unifies diffusion-based imitation learning and inverse reinforcement learning by learning a conservative energy field whose gradient drives action generation while exposing a recoverable reward signal, improving manipulation performanc… …
… Using only GRAIL-generated data, we train egocentric visual policies through a sim-to-real pipeline and deploy them on a Unitree G1 humanoid, achieving 84\% real-world success on diverse object pick-up and 90\% success on stair-climbing. …