How 800 Data Points Halve LLM Chain‑of‑Thought Length and Boost Accuracy

The ICLR‑2026 paper introduces LCPO, a lightweight preference‑optimization technique that uses only 800 curated examples and 50 training steps to cut large‑model chain‑of‑thought generation length by about 50% while maintaining or even improving answer accuracy, dramatically reducing training and inference costs.

Chain-of-ThoughtEfficient InferenceLCPO

0 likes · 8 min read

How 800 Data Points Halve LLM Chain‑of‑Thought Length and Boost Accuracy

JD Tech

Mar 12, 2025 · Artificial Intelligence

From Low‑Resource Large Model Training to Dynamic Margin Selection: A JD Engineer’s Journey

The article recounts a JD retail engineer’s rapid growth through tackling low‑resource large‑model training, developing a margin‑based dynamic data selection method (DynaMS) that earned an ICLR paper, and sharing practical insights on aligning business needs with cutting‑edge AI research.

AI researchICLRLow‑Resource Training

0 likes · 11 min read

From Low‑Resource Large Model Training to Dynamic Margin Selection: A JD Engineer’s Journey

JD Retail Technology

Mar 6, 2025 · Artificial Intelligence

Dynamic Margin Selection for Efficient Deep Learning and Low-Resource Large Model Training

Jia Xing’s research introduces Dynamic Margin Selection, a technique that repeatedly refreshes a core set of boundary‑close samples to train large language models efficiently on limited resources, achieving comparable loss to full‑data training, enabling six‑fold model compression, faster inference, and a proposed exponential scaling law for data‑efficient AI.

ICLRLarge Language ModelsLow‑Resource Training

0 likes · 10 min read

Dynamic Margin Selection for Efficient Deep Learning and Low-Resource Large Model Training