Author of the BaiMian large model, offering technology and industry insights.
The article explains how information entropy, cross‑entropy, and KL‑divergence shape reinforcement learning for large language models, describes the phenomenon of entropy collapse, compares token‑level and policy‑level entropy, and reviews recent methods like Clip‑Cov and KL‑Cov that mitigate this issue.
DeepSeek‑OCR introduces a novel visual encoder that transforms text into images, achieving up to 10‑20× token compression while maintaining OCR accuracy, and demonstrates strong performance on OmniDocBench with a 3B‑parameter model across multilingual and multimodal tasks.
This article breaks down the GPU memory requirements of large language models during training and inference, detailing the contributions of model weights, optimizer states, activations, KV cache, and activation recomputation, and provides concrete formulas, examples, and scaling insights for models like Qwen3 and DeepSeek V3.
LongCat-Flash-Thinking, the latest open‑source model from Meituan, introduces domain‑parallel RL training, a high‑throughput DORA infra, and a dual‑path inference framework that together achieve state‑of‑the‑art performance on logical, mathematical, coding, and agentic tasks while maintaining top‑tier speed.
This guide explains why naïvely adding special tokens during supervised fine‑tuning can destabilize a large language model, and provides step‑by‑step strategies—including tokenizer updates, embedding resizing, smart initialization, and LoRA‑based PEFT—to integrate new tokens while preserving the model's original capabilities.
A recent Hugging Face pull request reveals Alibaba’s upcoming Qwen3‑Next series, highlighting its extreme‑context, parameter‑efficient design that combines a 1:50 high‑sparsity MoE, a hybrid attention architecture mixing gated attention with Gated DeltaNet, and a Multi‑Token Prediction technique, promising ten‑fold throughput gains for 32K‑plus token contexts.
The article analyzes OpenAI's study on language‑model hallucinations, explaining how statistical limits in pre‑training and flawed binary evaluation incentives cause false answers, and proposes a confidence‑threshold scoring system that rewards honest "I don’t know" responses to improve reliability.
Atom-Searcher introduces an atomic‑thought reinforcement‑learning framework that decomposes complex reasoning into fine‑grained units, uses a Reasoning Reward Model to assign step‑wise rewards, dynamically balances process and result incentives, and achieves state‑of‑the‑art performance on multiple LLM benchmarks.
LongCat‑Flash is a 560‑billion‑parameter Mixture‑of‑Experts LLM that combines a dynamic zero‑computation expert design, shortcut‑connected MoE communication, variance‑aligned scaling, and a three‑stage agent‑centric pre‑training pipeline, delivering over 100 TPS on H800 GPUs at a cost of $0.70 per million tokens.
By replacing the traditional Pass@1 reward with a Pass@k formulation and a lightweight advantage computation, a 7B language model can dramatically improve its performance on math reasoning benchmarks, surpassing GPT‑4o while adding only a few lines of code and minimal training overhead.
