Tracked topic
…While Low-Rank Adaptation ( LoRA ) introduces additional weights between the LLM layers, Soft Prompting introduces additional fine-tuning-specific raw tokens to an LLM input. However, both require modification to the computational…
…To further address the sparsity of functional tokens during RL, we introduce Latent-Anchored GRPO (LA-GRPO), which stabilizes the training by anchoring functional tokens with a statically weighted auxiliary objective , providing…
…In the SFT stage, SlimSearcher employs Pareto-efficient filtration to distill trajectories that are both successful and economical, guiding the model toward inherently efficiency-aware search behaviors. During RL, we introduce Adaptive…
…Unlocking Vision-Language Alignment via Weight Recycling (2026) Learning Relative Representations for Fine-Grained Multimodal Alignment with Limited Data (2026) Unlocking Compositional Generalization in Continual Few-Shot Learning (2026) CrossFlowDG: Bridging the…
…https://github.com/A-EVO-Lab/a-evolve/tree/release/adaptive-auto-harness Get this paper in your agent: hf papers read 2606.01770 Don't have the latest CLI? curl -LsSf…
…Queries stay in floating point and are scored against the stored sketches. In the default 384-dimensional sentence-embedding setting, Clark Hash stores a cosine-search vector in 48 bytes instead of…
…It delivers a huge boost in compositional fidelity. 🔥 We’ve already open-sourced the dataset and model weights right here on Hugging Face. Check them out! 🎉 Super excited to share that this…
…hf papers read 2606.13674 Don't have the latest CLI? curl -LsSf https://hf.co/cli/install.sh | bash No model linking this paper Cite arxiv.org/abs/2606.13674 in…
…By comparing predicted judgments with ground-truth preferences and analyzing successes and failures in the reasoning process, the Orchestrator automatically refines its library of tools and skills without additional human annotation. Using…
…hf papers read 2605.30159 Don't have the latest CLI? curl -LsSf https://hf.co/cli/install.sh | bash No model linking this paper Cite arxiv.org/abs/2605.30159 in…