BestHub
Sign in
Tagged articles

post-training

31 articles · Page 1 of 1
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Aug 13, 2026 · Artificial Intelligence

Why RL Matters: From Reinforcement Learning to (Soft) Distillation

The article argues that reinforcement learning is crucial in post‑training because it refines and localizes chain‑of‑thought patterns learned during supervised fine‑tuning, improves model controllability, and can be complemented or substituted by distillation—especially soft distillation—to transfer high‑quality patterns from stronger teachers to weaker models.

LLMchain-of-thoughtdistillation
0 likes · 12 min read
Why RL Matters: From Reinforcement Learning to (Soft) Distillation
Machine Heart
Machine Heart
Aug 9, 2026 · Artificial Intelligence

Why Continual Learning Won’t Take Ten Years—Five Hot Paths and the Fight Against Catastrophic Forgetting

The article surveys five emerging approaches to LLM continual learning—external agent memory, context engineering, post‑training, pre‑training, and self‑modifying models—explaining how each tackles the core obstacle of catastrophic forgetting, citing benchmarks such as TRACE, ACE, and SDFT, and reflecting on Karpathy’s ten‑year timeline.

Agent MemoryContext EngineeringLLM
0 likes · 17 min read
Why Continual Learning Won’t Take Ten Years—Five Hot Paths and the Fight Against Catastrophic Forgetting
Model Perspective
Model Perspective
Jul 31, 2026 · Artificial Intelligence

Understanding the Post-Training Process in DeepSeek V4‑Flash

DeepSeek released the V4‑Flash model with the same architecture as the preview but a revamped post‑training pipeline—SFT, reinforcement learning with GRPO, and distillation—yielding dramatic benchmark jumps and illustrating how post‑training now defines the model's real‑world capabilities.

DeepSeekGRPOLLM training
0 likes · 11 min read
Understanding the Post-Training Process in DeepSeek V4‑Flash
AI Large-Model Wave and Transformation Guide
AI Large-Model Wave and Transformation Guide
Jul 31, 2026 · Artificial Intelligence

DeepSeek V4‑Flash Official Release: Agent Upgrade, Post‑Training Boost, and Codex Integration

DeepSeek announced the public beta of its V4‑Flash model, highlighting a dramatic agent capability upgrade, performance gains from post‑training that surpass the previous preview and rival Opus 4.8 on DSBench tests, native Responses API support, full Codex compatibility, and easy setup scripts for developers.

AI modelAgentCodex integration
0 likes · 6 min read
DeepSeek V4‑Flash Official Release: Agent Upgrade, Post‑Training Boost, and Codex Integration
DataFunSummit
DataFunSummit
Jul 24, 2026 · Industry Insights

Why High-Quality Data Is the New Bottleneck in Large Model Competition

In a four‑hour investor briefing, DeepSeek founder Liang Wenfeng explains that the real competitive edge for large language models now lies in the ability to continuously produce high‑quality training signals, a capability limited by time rather than capital.

AI industryData FlywheelDeepSeek
0 likes · 10 min read
Why High-Quality Data Is the New Bottleneck in Large Model Competition
Fighter's World
Fighter's World
Jun 21, 2026 · Artificial Intelligence

How Post‑Training Turns General AI into Enterprise‑Specific Intelligence

The article explains how post‑training shifts AI investment from massive pre‑training compute to targeted, low‑cost fine‑tuning that lets enterprises embed proprietary business logic, outlines the ROI formula based on data exclusivity, eval formalizability, and task frequency, and presents real‑world case studies.

AI economicsEnterprise AIevaluation
0 likes · 24 min read
How Post‑Training Turns General AI into Enterprise‑Specific Intelligence
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Jun 19, 2026 · Artificial Intelligence

Can Post‑Training Close the Gap to Mythos‑Level AI? Musk Says 9 Months, Tang Says Faster

The article analyzes whether post‑training on GLM‑5.1/5.2 can bridge the gap to the banned Mythos model, citing Musk’s nine‑month claim, Tang’s rebuttal, Mind Lab’s benchmark gains, architectural adaptations, and the high barriers that make post‑training a critical yet scarce capability in China.

AI developmentGLM-5.2IndexCache
0 likes · 9 min read
Can Post‑Training Close the Gap to Mythos‑Level AI? Musk Says 9 Months, Tang Says Faster
Machine Heart
Machine Heart
Jun 19, 2026 · Artificial Intelligence

Who Is Quietly Building China’s Mythos‑Level AI? Musk Says 9 Months, Tang Says It’s Not That Fast

The article analyzes China’s race to achieve Mythos‑level intelligence, contrasting Musk’s nine‑month claim with Tang’s skepticism, and highlights Mind Lab’s unique post‑training work on GLM‑5.1/5.2 that has already delivered significant benchmark gains, while outlining the technical hurdles and timeline uncertainties.

AI development in ChinaGLM-5.2Large Language Models
0 likes · 8 min read
Who Is Quietly Building China’s Mythos‑Level AI? Musk Says 9 Months, Tang Says It’s Not That Fast
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Jun 18, 2026 · Artificial Intelligence

Can a 3B Model Rival Claude Opus 4.5? Benchmark Gaps or Aggressive Post‑Training?

VibeThinker‑3B, a 3‑billion‑parameter language model built on Qwen2.5‑Coder‑3B, achieves scores within the range of 671 B‑parameter models on benchmarks such as LiveCodeBench, AIME26, IMO‑AnswerBench and GPQA, thanks to a two‑stage SFT, multi‑domain reinforcement learning, offline self‑distillation and a claim‑reliability (CLR) evaluator that together push its reasoning ability to the frontier.

Large Language ModelsParameter EfficiencyVibeThinker-3B
0 likes · 9 min read
Can a 3B Model Rival Claude Opus 4.5? Benchmark Gaps or Aggressive Post‑Training?
Network Intelligence Research Center (NIRC)
Network Intelligence Research Center (NIRC)
May 25, 2026 · Artificial Intelligence

What Does On-Policy Distillation Really Teach Large Language Models?

On-Policy Distillation (OPD) trains large language models by letting the student generate its own inference paths while the teacher supplies token‑level guidance, offering denser signals than RL but sometimes failing when teacher and student reasoning diverge, as detailed by THUNLP’s recent study.

Distillation MetricsLarge Language ModelsModel Alignment
0 likes · 8 min read
What Does On-Policy Distillation Really Teach Large Language Models?
Baobao Algorithm Notes
Baobao Algorithm Notes
May 22, 2026 · Artificial Intelligence

How LiteScale Cuts Wait Times in Large‑Model Post‑Training with Gradient Accumulation

The article examines the bottleneck of synchronous rollout in large‑model post‑training, proposes an asynchronous design using gradient accumulation and a global micro‑batch count to preserve loss equivalence, and introduces LogitsExpress for efficient top‑K knowledge‑distillation communication, all implemented in the lightweight LiteScale framework.

Distributed TrainingLarge Language Modelsasynchronous rollout
0 likes · 16 min read
How LiteScale Cuts Wait Times in Large‑Model Post‑Training with Gradient Accumulation
Machine Heart
Machine Heart
May 8, 2026 · Artificial Intelligence

How an 8B Video‑Language Model Beats GPT‑5 and Gemini‑3.1‑Pro at Cinematic Understanding

The CHAI framework introduced by CMU and Harvard defines a structured video‑language annotation scheme, scalable human‑AI oversight, and a post‑training pipeline that enables an 8B open‑source model to outperform closed‑source GPT‑5 and Gemini‑3.1‑Pro on professional cinematic techniques.

AnnotationQwen3-VLVideo Generation
0 likes · 11 min read
How an 8B Video‑Language Model Beats GPT‑5 and Gemini‑3.1‑Pro at Cinematic Understanding
Xiaomi Tech
Xiaomi Tech
Apr 27, 2026 · Artificial Intelligence

Xiaomi‑Robotics‑0: 20‑Hour Post‑Training Enables Seamless Earphone‑Box Assembly (Open‑Source)

The article details how Xiaomi‑Robotics‑0 achieves precise earphone‑to‑case insertion after only 20 hours of post‑training, outlines the sub‑millimetre precision challenges, presents a triple‑strategy (asynchronous execution, adaptive loss re‑weighting, Λ‑shape attention mask and random masking) to avoid the "lazy effect", and releases the full pipeline and code as open source for the robotics community.

Asynchronous ExecutionXiaomi Roboticsaction prefixing
0 likes · 6 min read
Xiaomi‑Robotics‑0: 20‑Hour Post‑Training Enables Seamless Earphone‑Box Assembly (Open‑Source)
Architect
Architect
Apr 25, 2026 · Artificial Intelligence

DeepSeek V4: 1M‑Token Context’s Impact on Model, Inference, Cache & Agents

The DeepSeek V4 technical report shows how a 1 million‑token context forces a redesign of attention, KV‑cache, optimizer, quantization and inference budgeting, turning long‑context capability from a costly showcase into a production‑ready feature for agents, search and Chinese professional tasks.

1M contextAgentic SearchAttention optimization
0 likes · 28 min read
DeepSeek V4: 1M‑Token Context’s Impact on Model, Inference, Cache & Agents
AIWalker
AIWalker
Apr 20, 2026 · Artificial Intelligence

How VA‑π Bridges Tokenizers and Autoregressive Generators for Pixel‑Perfect Images

VA‑π introduces a lightweight post‑training framework that uses variational inference and reinforcement learning to align tokenizers with visual autoregressive generators, achieving dramatic quality gains, extreme training efficiency, and robust pixel‑level reconstruction across diverse image generation tasks.

Autoregressive ModelsPixel Alignmentpost-training
0 likes · 14 min read
How VA‑π Bridges Tokenizers and Autoregressive Generators for Pixel‑Perfect Images
Machine Heart
Machine Heart
Apr 19, 2026 · Artificial Intelligence

World Engine: How Post‑Training Is Launching a New Era of Physical AGI

World Engine introduces a post‑training pipeline that combines high‑fidelity 3DGS simulation, hard‑case mining with diffusion generation, and reinforcement‑learning optimization to give autonomous‑driving models true decision‑making ability, surpassing data‑scaling limits and achieving significant safety gains in both industrial simulations and real‑world tests.

Simulationautonomous drivinghard case mining
0 likes · 11 min read
World Engine: How Post‑Training Is Launching a New Era of Physical AGI
Data Party THU
Data Party THU
Apr 12, 2026 · Artificial Intelligence

What’s Driving the Next Wave of LLM Post‑Training? A Deep Dive into SFT, RLHF, GRPO and Emerging Trends

This article systematically reviews the core post‑training techniques for large language models—including supervised fine‑tuning, RLHF, PPO, GRPO, DPO, RLVR and Agentic RL—explains their evolution, compares their trade‑offs, and highlights the most promising research directions for 2025‑2026.

AI AlignmentGRPOLLM
0 likes · 20 min read
What’s Driving the Next Wave of LLM Post‑Training? A Deep Dive into SFT, RLHF, GRPO and Emerging Trends
Machine Learning Algorithms & Natural Language Processing
Machine Learning Algorithms & Natural Language Processing
Mar 15, 2026 · Artificial Intelligence

Is RL Dead in LLM Post-Training? MIT’s RandOpt Challenges Traditional Methods

The MIT‑CSAIL paper introduces RandOpt, a single‑step, gradient‑free, fully parallel post‑training algorithm that adds Gaussian noise to pretrained LLM weights and ensembles the results, achieving or surpassing PPO/GRPO performance by exploiting dense "neural thickets" that emerge as model scale grows.

LLMRandOptensemble
0 likes · 12 min read
Is RL Dead in LLM Post-Training? MIT’s RandOpt Challenges Traditional Methods
PaperAgent
PaperAgent
Jan 8, 2026 · Artificial Intelligence

How SOP Enables Scalable Online Post-Training for Real‑World Robots

The SOP (Scalable Online Post‑training) framework redesigns VLA post‑training from offline, single‑machine, sequential processing to a distributed, parallel online system, allowing robot fleets to continuously learn, share experiences, and scale intelligence while maintaining stability and generalization in complex real‑world environments.

Distributed TrainingSOPScalable AI
0 likes · 11 min read
How SOP Enables Scalable Online Post-Training for Real‑World Robots
Baobao Algorithm Notes
Baobao Algorithm Notes
Nov 11, 2025 · Artificial Intelligence

Why Redesign the Training Stack? Inside Olmo‑Thinking’s Open‑Source RL Journey

This article provides a detailed technical analysis of the Olmo‑Thinking project, covering why a new open‑source LLM was built, the challenges of reinforcement learning at scale, data‑mix optimization, architectural bottlenecks such as missing GQA and QK‑Norm, and the post‑training techniques used to improve reasoning and long‑context capabilities.

Open-source ModelsRLVRdata selection
0 likes · 20 min read
Why Redesign the Training Stack? Inside Olmo‑Thinking’s Open‑Source RL Journey
Hailey Says
Hailey Says
Sep 7, 2025 · Artificial Intelligence

What to Consider When Discussing Large AI Models

The article explains how to choose suitable large AI models for product use by focusing on three key aspects—model inference, post‑training optimization techniques such as quantization, pruning and distillation, and the role of GPUs, including their kernels, software stack, and computing platforms.

GPUPruninginference
0 likes · 8 min read
What to Consider When Discussing Large AI Models
Alibaba Cloud Developer
Alibaba Cloud Developer
Jul 31, 2025 · Artificial Intelligence

Why Post‑Training Matters: Scaling Laws, Fine‑Tuning, and RL Strategies for LLMs

This article explores the importance of post‑training for large language models, explains scaling laws for pre‑ and post‑training, details common fine‑tuning methods (full, PEFT, LoRA), outlines alignment techniques such as RLHF, DPO, PPO, and presents practical workflows using Llama 3 and DeepSeek‑R1, while also discussing test‑time reasoning optimizations.

AlignmentFine-tuningLLM
0 likes · 19 min read
Why Post‑Training Matters: Scaling Laws, Fine‑Tuning, and RL Strategies for LLMs
Alibaba Cloud Big Data AI Platform
Alibaba Cloud Big Data AI Platform
Jun 25, 2025 · Artificial Intelligence

Boost Post‑Training Efficiency with Cosmos‑RL, Ray, and VeRL on Alibaba PAI

This article introduces Alibaba Cloud's PAI platform and demonstrates how open‑source reinforcement‑learning frameworks such as Cosmos‑RL, Ray, and VeRL accelerate post‑training for large language models, offering higher throughput, fault‑tolerance, and seamless integration for AI developers.

AI platformDistributed TrainingOpen Source Frameworks
0 likes · 9 min read
Boost Post‑Training Efficiency with Cosmos‑RL, Ray, and VeRL on Alibaba PAI
Baobao Algorithm Notes
Baobao Algorithm Notes
Mar 21, 2025 · Artificial Intelligence

Unlocking LLM Reasoning: A Deep Dive into Post‑Training Techniques

This article provides a comprehensive technical overview of large language model post‑training, covering fine‑tuning methods (full, parameter‑efficient, LoRA families, prompt tuning), domain‑adaptive tuning, reinforcement‑learning reward modeling, process vs. outcome rewards, inference‑enhancement strategies, dynamic compute allocation, verifier‑augmented reasoning, current challenges, and emerging research directions such as meta‑cognition, physical reasoning, and swarm intelligence.

LLMmeta-cognitionpost-training
0 likes · 21 min read
Unlocking LLM Reasoning: A Deep Dive into Post‑Training Techniques
Baobao Algorithm Notes
Baobao Algorithm Notes
Sep 29, 2024 · Artificial Intelligence

Decoding OpenAI o1: Test‑Time Scaling, PRM Search & Inference Strategies

This article analyses the training tricks behind OpenAI's o1 model, explaining test/inference‑time scaling laws, post‑training techniques, process‑supervised reward models (PRM), various inference‑time search methods, data‑collection pipelines, and the trade‑offs between allocating compute to pre‑training versus inference.

LLM inferenceOpenAI o1Reward Model
0 likes · 34 min read
Decoding OpenAI o1: Test‑Time Scaling, PRM Search & Inference Strategies