6 papers
SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD
Dongfang Li, Xiaodong Luo, Ruoyu Sun +64
Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pre…
Data Trajectory Alignment for LLM Domain Adaptation: A Two-Phase Synthesis Framework for Telecommunications Mathematics
Zhicheng Zhou, Jing Li, Suming Qiu +3
General-purpose large language models (LLMs) are increasingly deployed in verticals such as telecommunications, where adaptation is hindered by scarce, low-information-density corp…
Agentic-KGR: Co-evolutionary Knowledge Graph Construction through Multi-Agent Reinforcement Learning
Jing Li, Zhijie Sun, Zhicheng Zhou +4
Current knowledge-enhanced large language models (LLMs) rely on static, pre-constructed knowledge bases that suffer from coverage gaps and temporal obsolescence, limiting their eff…
Logits Replay + MoClip: Stabilized, Low-Cost Post-Training with Minimal Forgetting
Suming Qiu, Jing Li, Zhicheng Zhou +3
Large language models (LLMs) often face a trade-off in post-training: improvements on specialized domains frequently come at the expense of general capabilities. Existing solutions…
HES-SQL: Hybrid Reasoning for Efficient Text-to-SQL with Structural Skeleton Guidance
Suming Qiu, Jing Li, Zhicheng Zhou +3
We present HES-SQL, a novel hybrid training framework that advances Text-to-SQL generation through the integration of thinking-mode-fused supervised fine-tuning (SFT) with Group Re…
DeepJSONEval: Benchmarking Complex Nested JSON Data Mining for Large Language Models
Zhicheng Zhou, Jing Li, Suming Qiu +3
The internet is saturated with low-density, high-redundancy information, such as social media comments, repetitive news, and lengthy discussions, making it difficult to extract val…