9 papers
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
DeepSeek-AI, Anyi Xu, Bangcai Lin +315
We present a preview version of DeepSeek-V4 series, including two strong Mixture-of-Experts (MoE) language models -- DeepSeek-V4-Pro with 1.6T parameters (49B activated) and DeepSe…
Data-Efficient RLVR via Off-Policy Influence Guidance
Erle Zhu, Dazhi Jiang, Yuan Wang +8
Data selection is a critical aspect of Reinforcement Learning with Verifiable Rewards (RLVR) for enhancing the reasoning capabilities of large language models (LLMs). Current data…
MiniLLM: On-Policy Distillation of Large Language Models
Yuxian Gu, Li Dong, Furu Wei +1
Knowledge Distillation (KD) is a promising technique for reducing the high computational demand of large language models (LLMs). However, previous KD methods are primarily applied…
Trust-Region Adaptive Policy Optimization
Mingyu Su, Jian Guan, Yuxian Gu +2
Post-training methods, especially Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), play an important role in improving large language models' (LLMs) complex reasoning…
AgentBench: Evaluating LLMs as Agents
Xiao Liu, Hao Yu, Hanchen Zhang +19
The potential of Large Language Model (LLM) as agents has been widely acknowledged recently. Thus, there is an urgent need to quantitatively \textit{evaluate LLMs as agents} on cha…
Direct Preference Knowledge Distillation for Large Language Models
Yixing Li, Yuxian Gu, Li Dong +3
In the field of large language models (LLMs), Knowledge Distillation (KD) is a critical technique for transferring capabilities from teacher models to student models. However, exis…