128 upvotes · 13 JUL 2026 · Mikhail Komarov, Ivan Bondarenko, Stanislav Shtuka et al.
This paper introduces RAGU, an open-source GraphRAG engine that improves large language models with structured knowledge by separating extraction and consolidation, and trains a compact extractor that outperforms larger models on knowledge-graph construction and GraphRAG tasks. Practitioners might care because RAGU can efficiently generate more accurate and complete context for tasks like factoid-level evidence recall and multi-hop question answering.
30 upvotes · 11 JUL 2026 · Aleksandr Kutsakov, Mariia Sadovina, Georgii Gospodinov et al.
This paper develops a large audio language model that can answer questions with specific timestamps, improving its ability to understand long audio recordings. Practitioners in audio and speech recognition may care about this development as it enables more accurate and context-specific information retrieval from audio data.
24 upvotes · 18 JUL 2026 · Guangran Cheng, Chengqi Lyu, Songyang Gao et al.
This paper proposes a new method for reinforcement learning in large language models, called Group Entropy-Controlled Policy Optimization (GEPO), which helps balance exploration and exploitation by controlling entropy levels across different tasks. Practitioners might care about GEPO because it can lead to more balanced and task-specific exploration levels.
14 upvotes · 18 JUL 2026 · Seanie Lee, Sanjoy Chowdhury, Chao Jiang et al.
This paper proposes a method to generate synthetic data for training API-calling agents without the need for an actual environment. This is useful for scalability, as collecting high-quality data at scale can be difficult. Practitioners might care about this approach because it could speed up the development of AI agents that can interact with APIs.
9 upvotes · 20 JUL 2026 · Tianzhu Ye, Li Dong, Guanheng Chen et al.
This paper proposes a new way for large language models to learn from feedback, allowing them to retain more detailed information about the quality of their responses and learn from it in a more nuanced way. Practitioners might care because this approach could lead to better performance on tasks where the model doesn't have a clear way to evaluate its own output.
7 upvotes · 19 JUL 2026 · Chen Wang, Zhaochun Li, Jionghao Bai et al.
This paper proposes a new method called Distilled Reinforcement Learning that improves large language model post-training by providing fine-grained guidance to transfer new knowledge from a teacher model to a student model. Practitioners might care because it outperforms standard reinforcement learning and on-policy distillation methods in terms of knowledge transfer and model performance.