Hugging Face's logo

gg-hf-gg
community
Activity Feed

AI & ML interests

None defined yet.

Recent Activity
shirakiin  authored a paper about 7 hours ago
EmbeddingGemma 2: Unified and Efficient Multimodal Representations
aotrih  authored a paper 21 days ago
RVQ Position Aware Speculative Decoding for On Device Text to Speech
ZachNagengast  authored a paper 22 days ago
RVQ Position Aware Speculative Decoding for On Device Text to Speech
View all activity
sergiopaniego 
posted an update about 8 hours ago
sergiopaniego 
posted an update 1 day ago
sergiopaniego 
posted an update 2 days ago
view post
Post
114
Last week we closed the Training Agents series IRL at the GitHub offices in SF

4 talks, a panel and audience Q&A. almost 2h on post-training agents with open models

> Learnings from enterprise RL (Govind Kamtamneni Microsoft)
> Environments and benchmarks in industry (Victor Barres Mercor)
> How to train your agents with Miles (Tao Lin RadixArk)
> Why we built OpenEnv Arena (Xiangyi Li BenchFlow)

Full recording + Slides

https://www.youtube.com/watch?v=CKFUx7COIwQ
ZachNagengast 
authored a paper 22 days ago
EduardoPacheco 
authored a paper 22 days ago
sergiopaniego 
posted an update 23 days ago
sergiopaniego 
posted an update 23 days ago
view post
Post
299
Repo2RLEnv just shipped TaskSmith + 50 high quality RL envs generated from HF repos 🔨

TaskSmith is a specialized harness that turns a merged PR into a verified RL env

the envs come from HF repos (Transformers, TRL, PEFT, Accelerate, Diffusers), shipped as Harbor tasks you can eval or train on

> code: github.com/huggingface/Repo2RLEnv
> dataset: huggingface.co/datasets/FineEnvs/HF_ML_Tasksmith
sergiopaniego 
posted an update 24 days ago
view post
Post
3,822
ThinkingBox from @microsoft is now available as an OpenEnv env (cc @tuhink 🤗 )!

> ThinkingBox is a sandbox for testing agents on business workflows. it simulates a customer, gives the agent MCP tools over a real database, and at the end checks what changed in that database instead of trusting the agent's last message

> ThinkingBox-Bench is the benchmark built on it: 507 tasks across retail, insurance, travel, banking and consulting

> the OpenEnv env runs each task as an episode in its own isolated backend and returns a pass/fail reward from those checks

https://huggingface.co/blog/microsoft/thinkingbox
sergiopaniego 
posted an update 27 days ago
sergiopaniego 
posted an update about 1 month ago
view post
Post
463
ICYMI, Async GRPO in TRL now supports LoRA and we wrote a looong blog testing it

> the adapter is a few megabytes, so the weight sync is a file instead of an NCCL transfer
> 3 HF Jobs: 1 trainer and 2 vLLM replicas
> the adapter travels through an HF Storage Bucket mounted in all 3 at the same path
> a proxy in front of the replicas routes each rollout to the one already holding its KV prefix

https://huggingface.co/blog/asyncgrpo-lora-hfjobs
EduardoPacheco 
authored a paper about 1 month ago
sergiopaniego 
posted an update about 2 months ago
view post
Post
3,395
while preparing the last class of the Training Agents live series during the summer, i spent some time reading the post-training sections of many frontier model reports, to learn how they use RL environments to improve their models, and wrote a blog about it

if you use any kind of coding harness, or you saw the Blender scenes that went viral recently, this might be interesting to you

Blog: https://huggingface.co/blog/sergiopaniego/rl-environments-2026
srija616 
authored 4 papers about 2 months ago
sergiopaniego 
posted an update about 2 months ago
view post
Post
2,257
Can you do RL over taste?

I've spent some time reproducing, in the open, Surya N's idea of training a model to paint with code. It's a coding model that learns to paint watercolours by writing JS code, trained with GRPO. I used TRL and OpenEnv for this, with the whole pipeline running on Hugging Face.

The interesting part is that the reward has no correct answer, unlike a math problem. In this case it's based on the artistic preferences of the person who builds the dataset.

Everything is published: the environment, the reference pool, the trained adapters, every painting of every run with the code that made it, and a write-up with all the decisions, including the ones that went wrong.

Blog post: https://huggingface.co/blog/train-to-paint-with-code
sergiopaniego 
posted an update 2 months ago
view post
Post
613
catching up on some bookmarked reads from the summer, reading Antidoom from @liquidai

small reasoning models get stuck more easily when the task involves a long thinking trace and a hard problem. It starts repeating the same word over and over again ("Wait", "Alternatively"…), each repetition makes the next one likelier, and the generation is spent before it reaches an answer

they measured it, 10.2% of completions for an early LFM2.5-2.6B checkpoint and 22.9% for Qwen3.5-4B at greedy. After training those drop to 1.4% and 1.0%

the fix is FTPO (final token preference optimization). What I like is how narrow it is, it only touches the single token where the loop starts

three ways it differs from DPO:
> trains one token position, mid-generation, instead of whole sequences
> spreads probability across ~20 plausible alternatives instead of swapping one overtrained token for another
> keeps the regularizer in logit space, no softmax, so the rest of the vocabulary stays put

the third one is what makes it usable. If you want to edit one position without disturbing the model, you can't have a loss that reshuffles the other 150k logits on the way

and their explanation abt the result: the training teaches the model nothing new about math or code, it clears the failure mode that was blocking answers the model could already produce

full blog > https://www.liquid.ai/blog/antidoom

FTPO itself comes from Antislop, where it was built to strip overused phrasing. LiquidAI retargeted it to doom loops

and under the hood it's a subclass of TRL's DPOTrainer with compute_loss overridden, around 90 lines of loss and no new trainer

we documented that pattern in TRL's docs
https://huggingface.co/docs/trl/main/en/customization#change-the-training-objective
  • 1 reply
·
sergiopaniego 
posted an update 2 months ago
view post
Post
414
super interesting new paper from Microsoft "Agent Lightning v1.0: Towards Harnessed Agentic RL" by Zhiyuan He et al.

same idea we've seen already several times: you train the agent inside the real harness it ships with, instead of a reimplementation of it

now that recipe has a name → harnessed agentic RL

paper: huggingface.co/papers/2608.17528

the tricky bit they nail down: one rollout is not one training sample

the harness calls the model many times, so a single episode → a variable number of (prompt, response) rows

you don't even know the batch size until the episode finishes running

its real contribution is being first to systematically map the four problems that fall out of that:

> retokenization + sample merging
> advantage calculation over a variable sample count
> loss normalization at the rollout level, not per sample
> backend scheduling when the batch size is dynamic

and it actually works → plain RL inside the real harness, no reimplementation

Qwen3.5-9B on SWE-bench Verified 41.8 → 56.4 (+14.6), with only ~6k examples

the whole thing is ~3,500 lines, any harness, self-hosted k8s

from our side, we've shared some materials on the same line you may want to check out :)

> Agentic RL: Token-In, Token-Out Done Right: https://huggingface.co/blog/huggingface/tito
> a full worked example, opencode owning its loop trained with GRPO: https://huggingface.co/blog/sergiopaniego/trl-openenv-harness-training
> Harness, Scaffold, and the AI Agent Terms Worth Getting Right: https://huggingface.co/blog/agent-glossary

on a similar line:

https://x.com/SergioPaniego/status/2062911580564496576
henryNdubuaku 
authored a paper 2 months ago