…Expressive Performance Rendering from Deadpan Context Published on Jun 10 Submitted by Dmitry on Jun 12 Authors: Dmitrii Gavrilev Abstract PianoKontext generates variable-length piano performances by aligning MIDI scores with audio…
…Yihong Liu , , , Abstract COPSD transfers high-resource language model reasoning behavior to low-resource languages using self-distillation with crosslingual context, improving mathematical reasoning performance. AI-generated summary Large language models (LLMs…
…We also find that providing access to client profiles yields an average performance gain of 18.24\%, highlighting the importance of user-specific information for effective persuasion . Overall, our work highlights persona…
…AI-generated summary Parameter-efficient adaptation of pretrained vision models is commonly performed through linear probes, prompts, low-rank updates, or lightweight residual modules. While effective, these methods usually treat adaptation as…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct LLM agents increasingly retrieve externally curated skills-procedural instructions retrieved at decision time-to improve performance on long-horizon interactive tasks. Existing skill libraries…
…View arXiv page View PDF Project page GitHub 98 Add to collection Community Less number of 3D Gaussians, better performance Get this paper in your agent: hf papers read 2606.05102 Don…
…Across 11 image editing models, we find overall low performance, with the current highest-performing industry leader scoring only 17.1% (mIoU). Task decomposition reveals especially challenging operation types (geometric transformation, most…
…Our approach consistently improves LLM performance across multiple tasks, including instruction following and math reasoning, in both training free and fine tuning settings. Moreover, we show that our method mitigates attention sinks…
…Lei Yang , , Abstract Multi-domain reinforcement learning in language models causes performance degradation through shared computational pathways, but targeted refresh and rollback techniques can selectively recover lost capabilities with minimal side effects…
…Across these settings, κ- SwiGLU improves mean CORE performance while adding negligible parameters and incurring only a small computational overhead, demonstrating that confidence-aware gate sharpness is a promising mechanism for improving…