K2-V2 is a fully open, 360‑open large language model built from scratch, designed to serve as a superior base for reasoning adaptation while also supporting conversation and knowledge retrieval. It competes with leading open‑weight models in its size class, outperforming Qwen2.5‑72B and approaching Qwen3‑235B, and incorporates domain knowledge, reasoning, long‑context handling, and tool use throughout training. The authors release the complete training history, data composition, model weights, and LLM360 artifacts to enable community use and continuous training.
arXiv:2603.00889v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have exhibited remarkable reasoning capabilities, largely enabled by supervised fine-tuning (SFT)- and reinforce...
arXiv:2604. 10788v2 Announce Type: replace-cross Abstract: Tool-Integrated Reasoning (TIR) has emerged as a promising direction by extending Large Language Models' (LLMs) capabilities with external tools during reasoning.
InternBootcamp is an open‑source framework that offers over 1,000 domain‑diverse task environments for large language model (LLM) reasoning research. It introduces Bootcamp‑Eval, an automatically generated benchmark for comprehensive performance assessment. Experiments show that training on InternBootcamp significantly improves reasoning performance, with a 32B model achieving state‑of‑the‑art results on Bootcamp‑Eval and other established benchmarks, demonstrating that scaling the number of training tasks yields consistent gains.
arXiv:2509. 21013v4 Announce Type: replace-cross Abstract: Given the prohibitive cost of pre-training large language models, it is essential to leverage smaller proxy models to optimize datasets before scaling up.
arXiv:2608. 02975v1 Announce Type: cross Abstract: Large language models (LLMs) have demonstrated impressive performance in MQM-based translation quality (TQ) evaluation, and recent advances in large reasoning models (LRMs) promise even greater improvements.
Enrich-on-Graph (EoG) is a flexible framework that uses large language models to enrich knowledge graphs, thereby bridging the semantic gap between structured graphs and unstructured queries in complex reasoning tasks. By leveraging LLMs’ prior knowledge, EoG enables efficient evidence extraction from knowledge graphs, achieving precise and robust reasoning while maintaining low computational costs and scalability. The authors also introduce three graph quality evaluation metrics for query‑graph alignment, theoretically validate their optimization objectives, and demonstrate state‑of‑the‑art performance on two KGQA benchmark datasets.
arXiv:2604.02007v3 Announce Type: replace Abstract: Building general-purpose reasoning models using reinforcement learning with verifiable rewards (RLVR) across diverse domains has been widely adopte...
LongZero is a reinforcement learning method that enables large language models to improve long‑context reasoning without external data. It works by having the model first generate a compact, structured question‑answer pair with supporting evidence, then expanding that evidence into a long, multi‑document context. The approach combines self‑written multi‑hop reasoning and multi‑round retrieval tasks, trained with GRPO, and outperforms baseline RL methods on five long‑context reasoning benchmarks.
arXiv:2607. 14895v1 Announce Type: new Abstract: Reasoning language models (RLMs) have demonstrated impressive performance in domains such as mathematics and coding.
The paper investigates whether large language models (LLMs) truly grasp logical structure or merely rely on surface semantics. By evaluating models on logically equivalent inputs, the authors find that many LLMs fail to produce consistent predictions and struggle with unseen logical forms. They introduce LoGIST, a framework that maps inputs to decision diagrams and uses these logical embeddings to guide LLM reasoning, thereby reducing inconsistency and improving generalization.
The paper surveys efficient reasoning in large language models, contrasting fast intuitive (System 1) and slow deep (System 2) reasoning. It analyzes why System 2 is computationally costly yet more accurate, and why System 1 is efficient but less effective. The survey covers causes of inefficiency, patterns of reasoning behavior, and potential solutions to balance performance and computational budgets, offering actionable insights and an open‑source repository for ongoing research.
arXiv:2509. 03059v2 Announce Type: replace-cross Abstract: Recent advances in Large Language Models (LLMs) have shown that their reasoning capabilities can be significantly improved through Reinforcement Learning with Verifiable Reward (RLVR), particularly in domains like mathematics and programming, where ground-truth correctness can be automatically evaluated.