huggingface-projects/Deep-RL-Course-Certification
Viewer • Updated • 1.92k • 3.09k • 20
AI & ML interests
None defined yet.
Recent Activity
View all activity
sergiopaniego
updated a
dataset about 3 hours ago
sergiopaniego
posted an update about 18 hours ago
Post
1151
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
> 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
pcuenq
updated a
dataset 3 days ago
sergiopaniego
posted an update 4 days ago
Request to remove my entry from certified_users.csv
1
#4 opened 5 days ago
by 
Deepuamrita23115
Delete row per user request
#5 opened 4 days ago
by 
pcuenq
Post
222
I put together a Space to showcase tokenization-in-the-browser libs we maintain at HF
Let me know what you think!
sbrandeis/tokenizers-wasm-demo
Let me know what you think!
sbrandeis/tokenizers-wasm-demo
AdinaY
submitted a
paper to Daily Papers 14 days ago
sergiopaniego
posted an update 14 days ago
Post
368
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
> 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
fix: scope copy worker credentials to the Hub origin
#3 opened 14 days ago
by 
GuillaumeSalouHF
fix: scope copy worker credentials to the Hub origin
#2 opened 14 days ago
by GuillaumeSalouHF
sergiopaniego
posted an update 25 days ago
Post
3317
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
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
sergiopaniego
posted an update about 1 month ago
Post
2176
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
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 about 1 month ago
Post
558
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
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
sayakpaul
authored a
paper about 1 month ago
sayakpaul
submitted a
paper to Daily Papers about 1 month ago
sergiopaniego
posted an update about 2 months ago
Post
383
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
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
Failed to retrieve error logs: SSE is not enabled
1
#76 opened about 2 months ago
by 
javierrcyb
sergiopaniego
updated a
Space about 2 months ago
sergiopaniego
posted an update about 2 months ago
Post
796
Something I really like when I study a subject is understanding its history, how it reached the point where it is today
I did that exercise for RL in post-training: from RLHF and PPO, to verifiable rewards, to the GRPO family of variants, to agents acting in environments. Everything is backed by what the labs themselves say in their public reports (DeepSeek, Qwen, Kimi, GLM-5, Nemotron, Mistral and more), in their own words
This is the companion piece to Class 3 of our Training Agents series with @burtenshaw . The class explains how GRPO works, with three hands-on experiments. The article shows where the same ideas appear at frontier scale
https://huggingface.co/blog/sergiopaniego/agentic-rl-2026
I did that exercise for RL in post-training: from RLHF and PPO, to verifiable rewards, to the GRPO family of variants, to agents acting in environments. Everything is backed by what the labs themselves say in their public reports (DeepSeek, Qwen, Kimi, GLM-5, Nemotron, Mistral and more), in their own words
This is the companion piece to Class 3 of our Training Agents series with @burtenshaw . The class explains how GRPO works, with three hands-on experiments. The article shows where the same ideas appear at frontier scale
https://huggingface.co/blog/sergiopaniego/agentic-rl-2026
- 1 reply