长琴

... →

跳至内容
  • 首页
  • 导览
  • 存档
  • 拾文
  • 资源
  • 好玩
  • 观点
  • 关注
  • 关于
2026

OPD-RL 新范式:diff 出圈——从目标,到奖励,再到梯度

2026-10-15 •
Thinking

从Pass@k到advantage,一场意想不到的邂逅

2026-09-11 •
Thinking

「差值即信号」:OPD 蒸馏的新范式

2026-08-11 •
Feeling

SIS:不硬clip,大部分token其实可转为on-policy

2026-07-13 •
Feeling

从 SVM 到分布对齐:后训练方法的另一个统一视角

2026-07-04 •
Feeling

为什么TRPO在LLM里不能用?——FiberPO的起点

2026-06-22 •
Feeling

你可能没那么懂 SFT:SFT 与 RL 的爱恨纠葛

2026-06-01 •
Feeling

GAGPO:如果把GiGPO拉回PPO+GAE

2026-05-18 •
Feeling

TRPO深度拆解:为什么做后训练应该读懂TRPO

2026-05-11 •
Feeling

COPO:基于认知模式的 Step-Level Agentic RL 优化

2026-04-23 •
Feeling

LLM 的下一步:从“会答”到“会想”——Planning as Data 与思考范式重构

2026-04-17 •
Thinking

探寻实时学习新路径:挖掘极致高效的“子空间微调”

2026-04-11 •
Thinking

RL新范式:从经验到更高质量数据——我们不再训练模型,而是在制造数据

2026-03-29 •
Thinking

Training-Free RL:当“训练”不再更新参数,而是更新上下文

2026-03-24 •
Thinking

MoE RL 训练不稳定性再思考:训推不一致,还是采样噪声?

2026-01-22 •
Feeling

稳定压倒一切:MoE RL 训推不一致问题及解决策略

2026-01-17 •
Feeling

LLM 强化的“炼金术”:主流开源模型的 RL 优化策略赏析

2026-01-14 •
Thinking
2025

RL究竟能不能突破Base边界——关于推理能力外推、稳定性与训练条件的系统分析

2025-12-31 •
Thinking

Reward建模新范式:无验证RL——当模型只能相信自己,会发生什么?

2025-12-21 •
Thinking

DeepSeekV3.2后训练:稳定压倒一切

2025-12-03 •
Feeling

DeepSeekMath-V2自我验证:搞数据的风吹到了奖励模型

2025-11-29 •
Feeling

两处容易踩的坑:LLM 消息数组与字典工具的隐藏副作用

2025-11-23 •
Coding

Reward建模新范式:无验证器RL与Reference的妙用

2025-11-11 •
Thinking

重识LLM法则:上下文工程与数据进化

2025-07-27 •
Feeling

GiGPO:双层级优势函数驱动的Agent强化学习新范式

2025-07-25 •
Feeling

解锁模型潜能:Reward 数据如何塑造与激发 LLM 的推理策略

2025-07-13 •
Feeling

激活诱导LLM指令跟随

2025-07-01 •
Coding

指令跟随近期工作梳理(2025年上半年)

2025-06-26 •
Feeling

GRPO优化在继续——CISPO和熵

2025-06-19 •
Feeling

Reward Model建模

2025-06-09 •
Feeling

R1后范式最佳实践:Seed-Thinking和Qwen3

2025-05-01 •
Feeling

Yarz-Logic:R1-Zero相关实验报告

2025-04-26 •
Feeling

VAPO:基于价值方法的新突破

2025-04-19 •
Feeling

R1相关:R1-Zero的进一步理解和探索

2025-04-10 •
Feeling

异曲同工之妙的DrGRPO——DAPO几乎同时出现的又一GRPO优化!

2025-03-28 •
Feeling

DAPO:为GRPO的锦上加四点花

2025-03-19 •
Feeling

DeepSeek R1后LLM新范式

2025-03-15 •
Thinking

R1相关:DPO数据选择与DPO等RL算法

2025-03-02 •
Feeling

预训练:NTP和Scaling Law

2025-02-28 •
Feeling

LLM、强化、蒸馏讨论

2025-02-27 •
Feeling

R1相关:RL数据选择与Scaling

2025-02-27 •
Feeling

R1相关:少量高质量数据SFT激活LLM推理能力

2025-02-18 •
Feeling

DeepSeek R1深度技术解析及其影响

2025-02-17 •
Thinking

预训练:无处安放的躁动之心

2025-01-05 •
Feeling
2024

MIO

2024-09-28 •
Feeling

Tiny LLM Continual Pre-training:RHO-1

2024-04-13 •
Feeling

LLM中的演绎推理、归纳推理和溯因推理

2024-04-06 •
Feeling

LLM Tiny Pretrain:H2O-Danube and Stable LM

2024-02-03 •
Feeling

LLM DataManagement:Weaver

2024-02-01 •
Feeling

LLM DataManagement:Ziya2

2024-01-29 •
Feeling

LLM Continual Pre-training:Ziya2

2024-01-23 •
Feeling
2023

关于大语言模型的思考

2023-10-15 •
Thinking

ChatGPT 开发指南:Hugging LLM Hugging Future

2023-04-22 •
Thinking

ChatGPT 基础科普:知其一点所以然

2023-04-15 •
Thinking

ChatGPT 影响冲击:职业、行业与产业

2023-02-21 •
Feeling

ChatGPT 标注指南:任务、数据与规范

2023-02-19 •
Feeling

ChatGPT Prompt 示例

2023-01-31 •
Feeling

语言模型级联

2023-01-27 •
Feeling

ChatGPT Prompt工程:设计、实践与思考

2023-01-25 •
Thinking

Put Human in NLP Loop

2023-01-21 •
Feeling

GPT3 和它的 In-Context Learning

2023-01-20 •
Thinking
2022

W2NER解读

2022-10-30 •
Feeling

Global Pointer:Novel Efficient Span-based Approach for NER

2022-10-16 •
Feeling

DeepGen:Diverse Search Ad Generation and Real-Time Customization

2022-10-15 •
Feeling

FLAN:Fine-tuned Language Models are Zero-Shot Learners

2022-08-28 •
Feeling

W2NER 代码

2022-07-17 •
Coding

统一NER为词词关系分类

2022-06-11 •
Feeling

预训练模型与传统方法在排序上有啥不同?

2022-04-23 •
Feeling

MarkBERT

2022-04-23 •
Feeling

量化NLM模型的记忆力

2022-04-15 •
Feeling

NLP预训练模型的不可能三角

2022-04-15 •
Feeling

Training Data is More Valuable than You Think

2022-04-05 •
Feeling

句子表征综述

2022-03-27 •
Thinking

T5:Exploring the limits of Transfer Learning with a Unified Text-to-Text Transformer

2022-03-05 •
Feeling

ExT5:Towards Extreme Multi-Task Scaling for Transfer Learning

2022-02-19 •
Feeling
2021

Pretrain, Prompt and Predict, A Systematic Survey of Prompting Methods in NLP

2021-12-04 •
Feeling

Data Augmentation Approaches in Natural Language Processing:A Survey

2021-11-28 •
Feeling

Debiasing Techniques for Pre-Trained Language Models

2021-11-18 •
Feeling

浅析文本分类——情感分析与自然语言处理

2021-10-27 •
Coding

TensorBay 指南

2021-09-21 •
Coding

R-Drop

2021-08-18 •
Feeling

UniLM

2021-07-31 •
Feeling

通过最优转移进行词表学习:VOLT

2021-07-18 •
Feeling

简单的对比学习框架:SimCSE

2021-07-10 •
Feeling

预训练模型的过去、现在和未来

2021-06-20 •
Feeling

对NLP预训练模型的思考

2021-06-10 •
Thinking

Few-Shot NER and BERT Noisy Learning:ProtoBERT Paper Note

2021-06-06 •
Feeling

深度探索 Bert:BERTology Paper Note

2021-05-22 •
Feeling

SqueezeBERT 论文笔记

2021-01-17 •
Feeling
2020

从 Sentence-BERT 谈句子表征

2020-12-27 •
Feeling

Bert-Flow 论文笔记

2020-12-13 •
Feeling

NLP 表征的历史与未来

2020-12-12 •
Thinking

Funnel Transformer 论文笔记

2020-10-13 •
Feeling

PEGASUS 论文笔记

2020-09-13 •
Feeling

DeBERTa 论文+代码笔记

2020-06-27 •
Feeling

RoBERTa 论文+代码笔记

2020-06-25 •
Feeling

Bart 论文+代码笔记

2020-06-13 •
Feeling

中文分词系列一:思考分词

2020-05-13 •
Thinking

ALBERT 论文+代码笔记

2020-05-10 •
Feeling

DistilBERT 论文+代码笔记

2020-04-27 •
Feeling

Transformer 代码笔记

2020-04-23 •
Feeling

Luong Attention 论文+代码笔记

2020-04-14 •
Feeling

GPT-2 论文+代码笔记

2020-04-07 •
Feeling

Node2Vec 论文+代码笔记

2020-03-30 •
Feeling

TextRank Keyword Extraction 论文+代码笔记

2020-03-21 •
Feeling

Reformer, The Efficient Transformer 论文笔记

2020-02-15 •
Feeling

Bahdanau Attention 论文笔记

2020-02-08 •
Feeling

Encoder-Decoder Models Attention and Contextual Embedding Note (SLP Ch10)

2020-01-23 •
Feeling
2019

Neural Architectures for Named Entity Recognition 论文笔记

2019-12-28 •
Feeling

Few-Shot Charge Prediction with Discriminative Legal Attributes 论文笔记

2019-12-15 •
Feeling

关系提取简述

2019-12-11 •
Coding

ELECTRA 论文笔记

2019-12-08 •
Feeling

自然语言记忆模块(NLM)

2019-12-02 •
Thinking

CTRL 论文+代码+实践笔记

2019-09-28 •
Feeling

Bert 论文笔记

2019-08-05 •
Feeling

Transformer 论文笔记

2019-08-04 •
Feeling

ERNIE Tutorial(论文笔记 + 实践指南)

2019-08-02 •
Feeling

ChatBot 设计方案

2019-07-20 •
Thinking

Statistical Parsing Note (SLP Ch14)

2019-07-17 •
Feeling

XLNet 论文笔记

2019-07-14 •
Feeling

Syntactic Parsing Note (SLP Ch13)

2019-07-08 •
Feeling

Formal Grammars of English Note (SLP Ch12)

2019-06-21 •
Feeling

Sequence Processing with Recurrent Networks Note (SLP Ch09)

2019-06-17 •
Feeling

Part-of-Speech Tagging Note (SLP Ch08)

2019-06-11 •
Feeling

Neural Networks and Neural Language Models Note (SLP Ch07)

2019-05-31 •
Feeling

Vector Semantics Note (SLP Ch06)

2019-05-16 •
Feeling

Logistic Regression Note (SLP Ch05)

2019-05-08 •
Feeling

Naive Bayes and Sentiment Classification Note (SLP Ch04)

2019-05-05 •
Feeling

Regular Expressions, Text Normalization, and Edit Distance Note (SLP Ch02)

2019-04-22 •
Feeling

Information Extraction Note (SLP Ch18)

2019-04-09 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch18)

2019-04-08 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch17)

2019-04-08 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch16)

2019-04-04 •
Feeling

《纳博科夫最喜欢的词》读书笔记与思考

2019-03-31 •
Thinking

自然语言计算机形式分析的理论与方法笔记(Ch15)

2019-03-29 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch14)

2019-03-22 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch13)

2019-03-15 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch12)

2019-03-11 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch11)

2019-03-01 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch10)

2019-02-27 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch09)

2019-02-21 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch08)

2019-02-15 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch07)

2019-01-31 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch06)

2019-01-18 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch05)

2019-01-15 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch04)

2019-01-09 •
Feeling
2018

自然语言计算机形式分析的理论与方法笔记(Ch03)

2018-12-22 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch02)

2018-10-11 •
Feeling

自然语言计算机形式分析的理论与方法笔记(Ch01)

2018-09-19 •
Feeling

NLP 与 AI

2018-07-22 •
Thinking

2018CCF-GAIR:自然语言如何商业落地摘录及思考

2018-07-12 •
Thinking
2017

Language Model Note (SLP Ch03)

2017-10-14 •
Feeling

语言、AI、情感

2017-09-07 •
Thinking

自然语言处理(NLP)知识地图

2017-04-09 •
Feeling

Recents

  • OmniDreams:实时世界模型背后的 Self Forcing 与 DMD
  • 想做,但可以不做:AI 时代的注意力与自由
  • OPD-RL 新范式:diff 出圈——从目标,到奖励,再到梯度
  • 良渚大屋顶半日闲——阅读与休憩
  • 从Pass@k到advantage,一场意想不到的邂逅

Categories

  • Coding78
  • Feeling170
  • Thinking68

Tag Cloud

AAPO ACT AE AGAPO AGI AI AI-Coding AIGC ALBERT AR ASPO AUC Accuracy Activation Activation Steering Adam Age Agent Agentic-RL AgenticRL Aha Algorithm Array Arrow Attention Automatic Speech Processing Automation BAPO BERT BIO BIOHD BM25 BPE BabyGrow Backtracking Backward Bahdanau Attention Bart Bayes Beam Search Bert-Flow Bi-LSTM Biasing BigCodec Binary Search Blending Book Brain Brain Decoding Bridge Business C C.AI C4 CARD CCG CE BERT CFG CISPO CKY CNN COPO CRF CRL CS CYK Calculus Camera Cascades Catalan ChatBot ChatGPT Chi2 Chunking Class Imbalance Loss Classification Clip CoT Codec Cognition Collaborative Filtering Collins Parser CompoundEngineering Computational Linguistics Computer Computer Science Confusing Labels Context Engineering Context Learning Context-Free Grammars Continual Pre-training Continual Pretraining Continuous Learning Contrastive-Learning Coordinate Ascent Cosine Cosine Similarity Cross Entropy Cross-brackets Cross-view Ctrl Culture DA DAC DAPO DB DCPO DELTA DLM DMD DNN DP DPO Daily Darling Data Augmentation Data Clearing Data Enhancement Data Preprocess Data Science Data Structure DataManagement Database DeBERTa Debiasing Decoder Decoding Deep DeepGen DeepGraph DeepLearning DeepScaleR DeepSeek DeepSeek-GRM DeepSeek-V3.2 DeepSeekMath-V2 DeltaNet Dependence Diary Direct-OPD Disentangled Attention DistilBERT Distillation Django Docker Docker-Compose Dockerfile Dr GRPO DrDAPO DrGRPO Dream Dropout Dynamic-Mask EDA EM EMD EMPG EMPO ERL ERNIE ETTRL EVOL-RL EXAONE Economics Edit Distance Efficient-DeepLearning Elasticsearch Electra Elixir Ellipsis Embedding Embeddings Embodied Embodied AI EmbodiedAI Encoder Entropy Evaluation Eventlet ExT5 Exam F1 FD Leak FDW FLAN FSM Faith FastCuRL Feature Engineering Feature-based Few-Shot Few-shot Prompting FiberPO Fine-tuning Flash-Attention Formal Grammars Forward Freedom Full-Text-Search Function Syntax Funk MF Funnel Transformer Future GAE GAGPO GBTD GELU GFT GLU GMPO GP GPT-1 GPT-2 GPT-3 GPT3 GPU GRM GRPO GRPOEA GRU GSG GSPO GTPO GTPO-S Gan Garden-path Gated DeltaNet GiGPO Git Global Pointer Glow Graceful Shutdown Gradient Descent Graph GraphQL Grid Grammar Growth H2O-Danube HMM HPT Hard-SVM Hinge Loss Hope Host-only HuggingLLM Human-Attention Human-in-Loop Human-in-the-Loop ICL IDE IE IQR IcePop Imbalance Data Impossible-Triangle In-Context Learning Industry Inference Inference Scaling Information Extraction Information Theory Instruct InstructGPT Instruction Following Instruction Inference Intuitor Isolation Forest ItemCF JEPA Jaccard Java Jax Job Jupyter JustGRPO K2 KAT KKT KL KS Kernel Kernel Function Kernel Method Keyword Kimi Knowledge Graph L1 LCPO LIMD LIMO LIMR LLM LLM-Colosseum LLaDA LM LOF LR LSTM Labeling Language Model LayerNorm Lexical Semantics Lexicalism Lexicalized CFG Lexicalized Grammars Life Lightning OPD Linear Algebra Linear Sturcture Linked List LinkedList Linux Listen Llama LoRA LoRA-XS Real-time Learning Logistic Regression Lucene Luong Attention MDLM MEMM MF MIO MM Fusion MOPD MR-Search MTL Machine Machine Learning Machine Translation Manacher Managemnt MarkBERT Markov Materialized Views Math Matplotlib Matrix Factorization Median MemAPO Meta Learning Meta RL Metric MiCA MiMo MiniMax Minimum Edit Distance Minkowski MoE Model Evaluation Module Monad Monkey Patch Multi-Head Attention Multi-Modal MultiModal Multitask Multiway Tree NAT NER NLG NLM NLP NLU NMT NNW NOVER NTP Naive Bayes Neo4j Network Ngram NodeJS Normalizing Flow NumPy Numba Numpy OEL OMNI OPD OPRD OPSD ORZ Occupation OmniDreams One-Shot Online Learning Online Softmax Online-DPO-R1 OpenAI OpenClaw OpenSource OpenSpec Orientation P-R PCCG PCFG PEFT PEGASUS PLM PPMI PPO PTM PageRank Palindromic Pandarallel Pandas Partial Parsing Pass@k Passion Pearson Philosophy Phrase Structure Grammar Phrase Structure Grammars Physics Planning PoS Polars Poly-OPD Pooling Position-Encoding Post-Training Post-training Postgres Pragmatic Automatic Processing Pre-Trained Pre-Training Pre-training Precision Pretrain Pretrained Pretraining ProGRPO Probabilistic Grammar Probabilistic Model Promote Prompt ProtoBERT Pruning Psychology PyPI Python QA Quant Quantization Query Queue Qwen Qwen3 Qwen3-Next R-Drop R1 R1-Zero R3 RAG RAVR REER RELU RENT RESTRAIN RFE RGR RHO RHO-1 RL RLHF RM RM-R1 RMSE RMSProp RNN ROC RP-OPSD RWD Rank RaspberryPi Raspberrypi Reading Reasoning Recall Recommendation Recursion Reduction Reformer Regex Regular Expression Reinforce++ Reinforcement Learning Relationship Extraction Representation Reqular Expressions Retrieving Reward Reward Hack RoBERTa RolePlay Rotated Sorted Array Rust SAPO SCFG SFT SGD SIS SLM SMO SQL SRN SRT STAR-LDM STEP-OPD STaR SVD++ SVM Scaling Scaling Law Seaborn Search Seed-Thinking Segmentation Selection-Inference Self Forcing Self-Attention Self-Verified Semantic Automatic Processing Semantic Similarity Senta Sentence Representation Sentence Similarity Sentence-BERT Sentiment Classification SentimentAnalysis Sentry Siamese Sigmoid SimCSE Similarity Simon Simple-Zoo Simpson Paradox Skill Skywork Reward Slide Smoothing Soft-SVM Softmax Sort Span Sparse Attention Spell Check Spurious Reward SqueezeBERT Stability Stable LM Stack Stacking Statistics Stirling Strategic StratifiedKFold Streaming String Study Style Substring Summarization Supertagging Swap System T5 TF-IDF THW TIS TRPO TRT TS3-Codec TTRL TTS Tagging Talkie TanH TensorBay Tensorflow Test Text Classification Text Generation Text Normalization TextCNN TextRank Thinking Thought TiDAR TinyLoRA Tokenizer Transformer Transformer-XL Tree Treebank Tuning Tutorial Ubuntu UniLM Unity Operation Unix Unsupervised Elicitation UserCF VAD VAPO VITS VLA Vagrant Valence Vector Semantics Verifier Virtual Network VirtualBox Visualization Viterbi Vocabulary Learning VoiceAgent Voila Voting W2NER W2S OPD WOE Web Server Multithreaded Server Wide Word2vec Work World Model XLNet XTTS Z-Score Zero-Short Zero-Shot Zero-shot ZhouZhihua Zipf Ziya antigravity attention sink bias binning context emacs few-shot ffmpeg gated attention gpt-oss harmony format jpype kanban knowledge Graph lightinfer motion node2vec oat-zero off-by-one attention orz pararun promptlog s1 skill spec ssh str trae vim vlc

Music

© 2026 hscspring All rights reserved.

Powered by Hexo

搜索

输入关键词开始搜索