Topic
Training & post-training
Pre-training, reinforcement learning, fine-tuning and synthetic data, and the frameworks and environments for them.
Follow this topic
On this site · ElsewhereTopic graph
16 nodes · 47 links- Entries in this topic
- Neighbours outside the topic
- Builds on · Successor · Extends
- Prerequisite · Uses
- Explains · Supports
- Related
Skills and knowledge entries are written in Chinese, as are signals from Chinese sources.
Published signals
12 signals记忆张量开源 MemT 1.0：把记忆原生写进 Transformer
记忆张量开源记忆原生模型 MemT 1.0，含 36B 总参数、激活 3B 的 Nano 与 288B 总参数、激活 17B 的 Flash 两个型号，在 Transformer 模块中加入与注意力并行的记忆分支，把经验编码为可读写的内部状态。论文称这是记忆原生模型首次完成近 300B 参数规模的工程验证。
TorchNeuron: running HF models on AWS Trainium is easier than ever
TorchNeuron: running HF models on AWS Trainium is easier than ever
MiMo-V2.6 技术报告：小米花 260 万美元跑一次强化学习
小米 MiMo 团队 10 月 8 日发布 MiMo-V2.6 技术报告：Pro（1.02T 总参数、激活 42B）与 Flash（310B、激活 15B）两款 MoE 模型，强化学习同时扩大批量、任务环境与评分算力。每步约 2.5 万条轨迹、27 亿至 37 亿 token，Pro 的强化学习花费约 260 万美元，其中评分器占 12.7%。训练全程冻结 MoE 路由，作弊轨迹控制在 2% 以下；训练动态、强化学习环境与框架开源。
Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses
Training AI agents with reinforcement learning can be challenging because their tools, context, and decision-making are managed by complex frameworks. Agent Lightning connects existing agents to RL training, making it easier to improve them without rebuilding…
Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs
Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs
DeepSeek DSec：智能体训练拼的是沙盒
DeepSeek 公开智能体训练用的沙盒平台 DSec：单个生产单元约 160 个节点，每天约 300 万个沙盒，并发 38 万个以上，每秒创建 5,000 个以上。它用一套接口提供四种后端——无状态函数调用、容器、微虚拟机、完整虚拟机——按任务的隔离要求分档，并与强化学习框架协同设计，把有状态的推演和可被抢占的 GPU 训练拆开。训练环境第一次被一家实验室当成一等工程对象完整写出来。
Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic
Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal: refusing the right subset of a topic rather than the whole topic.
BigBang-v1：把数据生产系统本身当成优化对象，以及一处对不上的口径
上海交大人工智能学院、深势科技与上海算法创新研究院组成的无尽前沿团队发布 BigBang-v1：35B 参数、推理时约 3B 激活、262K 上下文，后训练数据称 100% 由 AI 自主合成，Apache-2.0 开源。可复用的不是跑分，而是它对任务的两个硬条件——必须同时前沿且可验证；缺一，合成数据都会迅速贬值。附一条核对：报道称它是「首个 RSI 原生训练的基座模型」，而 Hugging Face 上的元数据把它标为 Qwen3.6-35B-A3B 的微调。
Deploy local agents everywhere with LFM2.5-2.6B
LFM2.5-2.6B is built to power capable agents entirely on-device. It supports tool calling and multi-step workflows while staying small and fast enough for everyday hardware, from laptops to phones.
Fine-tune video and image models at scale with NVIDIA NeMo Automodel and Diffusers
NVIDIA and Hugging Face ship a NeMo Automodel integration that brings production-grade distributed diffusion training to any Diffusers-format model on the Hub — no checkpoint conversion, no model rewrites, scaling from one GPU to hundreds, Apache 2.0.
GPT-Red: Unlocking Self-Improvement for Robustness
OpenAI introduces GPT-Red, an automated red-teaming system that uses self-play to continuously probe and harden models against safety, alignment and prompt-injection failures.
Data for Agents
NVIDIA argues that agentic AI needs open data — software engineering traces, tool-use failures, multi-step reasoning, retrieval, safety and user simulation — and that synthetic data is how it scales, backed by the Nemotron open dataset family.