Reinforcement Learning for Code Optimization
Paper • 2607.25970 • Published • 14
AI & ML interests
None defined yet.
Recent Activity
View all activity
Fix: support reasoning_effort in chat template
1
#4 opened 14 days ago
by 
Molbap
Fix: support reasoning_effort in chat template
👍 1
#65 opened 14 days ago
by Molbap
pierrechambon
submitted a
paper to Daily Papers 2 months ago
ccsasuke
authored 9
papers 3 months ago
Two-Level Meta-Rubrics for Evaluating Open-Ended Generation: GAMUT, a Benchmark for Factual Completeness
Paper • 2607.19322 • Published • 11
Knowledge Extraction on Semi-Structured Content: Does It Remain Relevant for Question Answering in the Era of LLMs?
Paper • 2509.25107 • Published
SCRIBES: Web-Scale Script-Based Semi-Structured Data Extraction with Reinforcement Learning
Paper • 2510.01832 • Published
Learning Facts at Scale with Active Reading
Paper • 2508.09494 • Published
Learning to Reason for Factuality
Paper • 2508.05618 • Published • 7
FACTORY: A Challenging Human-Verified Prompt Set for Long-Form Factuality
Paper • 2508.00109 • Published • 4
Scene-LLM: Extending Language Model for 3D Visual Understanding and Reasoning
Paper • 2403.11401 • Published
Hierarchical Video-Moment Retrieval and Step-Captioning
Paper • 2303.16406 • Published
Post-training an LLM for RAG? Train on Self-Generated Demonstrations
Paper • 2502.10596 • Published
ccsasuke
submitted a
paper to Daily Papers 3 months ago
derenlei
authored 2
papers 4 months ago
qgallouedec
posted an update 5 months ago
Post
12337
Shipped hf-sandbox! 🥡
🧪 Running an eval that executes model-generated C on a few thousand prompts? You probably don't want any of that on your laptop.
Just shipped hf-sandbox, a Modal-style sandbox API on top of Hugging Face Jobs. Spin up an isolated, ephemeral container, run untrusted code, get the result back. No Docker on your laptop, no infra to manage.
Just pip install hf-sandbox.
Early days (v0.1); feedback and issues very welcome:
👉 https://github.com/huggingface/hf-sandbox
🧪 Running an eval that executes model-generated C on a few thousand prompts? You probably don't want any of that on your laptop.
Just shipped hf-sandbox, a Modal-style sandbox API on top of Hugging Face Jobs. Spin up an isolated, ephemeral container, run untrusted code, get the result back. No Docker on your laptop, no infra to manage.
Just pip install hf-sandbox.
Early days (v0.1); feedback and issues very welcome:
👉 https://github.com/huggingface/hf-sandbox
- 1 reply
qgallouedec
posted an update 5 months ago
Post
1689
**TRL v1.4 is out 🚀** Chunked NLL loss for SFT and a first-class **OpenReward** integration.
**Chunked NLL loss for SFT — drops peak VRAM by up to 14×**
Standard SFT materializes a full
Peak GPU memory, AdamW fp32:
- Qwen3-14B, 8×H100 FSDP2, 16k seq: 58.9 GB → 38.9 GB
- Qwen3-4B, 1×H100 80GB, 16k seq: OOM → 63.8 GB
- Qwen3-32B, 8×H100 FSDP2, 8k seq: OOM → 71.2 GB
End-to-end it's consistently as fast or faster than
Works with PEFT and VLMs out of the box.
**Open Reward Standard environment adapter**
The new
v1.4 also brings MFU helpers for dense + MoE models, GRPO support for Liger 0.8.0 (delta clipping + VESPO + KL bias correction), Tülu 3's length-normalized DPO loss, four more training chat templates (Cohere, Cohere2, Gemma 3, Qwen3-2507), and a 5+ GB CUDA memory leak fix in activation offloading.
Full release notes: https://github.com/huggingface/trl/releases/tag/v1.4.0
**Chunked NLL loss for SFT — drops peak VRAM by up to 14×**
Standard SFT materializes a full
[batch × seq × vocab] logits tensor before computing cross-entropy, which dominates peak memory at long context lengths. The new loss_type="chunked_nll" path drops ignored-label tokens before the lm_head matmul and computes cross-entropy in checkpointed chunks of 256.Peak GPU memory, AdamW fp32:
- Qwen3-14B, 8×H100 FSDP2, 16k seq: 58.9 GB → 38.9 GB
- Qwen3-4B, 1×H100 80GB, 16k seq: OOM → 63.8 GB
- Qwen3-32B, 8×H100 FSDP2, 8k seq: OOM → 71.2 GB
End-to-end it's consistently as fast or faster than
nll, and unlocks sequence lengths that don't fit at all under the standard path.SFTConfig(loss_type="chunked_nll")Works with PEFT and VLMs out of the box.
**Open Reward Standard environment adapter**
The new
trl.experimental.openreward adapter plugs any environment speaking the [Open Reward Standard](https://openrewardstandard.io) protocol into any TRL trainer that takes an environment_factory. One string — a catalog name or a URL — wires the dataset, factory, and reward_func slots; tools are bound dynamically from JSON Schema, no per-env wrapper code:from trl import GRPOTrainer
from trl.experimental.openreward import OpenRewardSpec
spec = OpenRewardSpec("Eigent/SETA", num_tasks=64)
trainer = GRPOTrainer(
...,
train_dataset=spec.train_dataset,
environment_factory=spec.environment_factory,
reward_funcs=spec.reward_funcs,
)v1.4 also brings MFU helpers for dense + MoE models, GRPO support for Liger 0.8.0 (delta clipping + VESPO + KL bias correction), Tülu 3's length-normalized DPO loss, four more training chat templates (Cohere, Cohere2, Gemma 3, Qwen3-2507), and a 5+ GB CUDA memory leak fix in activation offloading.
Full release notes: https://github.com/huggingface/trl/releases/tag/v1.4.0
qgallouedec
posted an update 5 months ago
Post
9205
TRL v1.3 ships day-one training support for Qwen 3.6 🚀
The new Qwen 3.6 family (
Qwen/Qwen3.6-27B, Qwen/Qwen3.6-35B-A3B) reuses the Qwen3.5-MoE architecture but ships a slightly different chat template, so we updated the stack end-to-end: new training template with {% generation %} markers, tool-call response schema routing, tiny test models for the VLM matrix.SFT with assistant-only loss works out of the box:
from trl import SFTConfig, SFTTrainer
trainer = SFTTrainer(
model="Qwen/Qwen3.6-27B",
args=SFTConfig(assistant_only_loss=True),
train_dataset=dataset,
)
trainer.train()So does GRPO tool-calling — just hand
tools=[...] to GRPOTrainer.v1.3 also brings a new experimental TPO trainer (Triple Preference Optimization), speculative decoding in
trl vllm-serve (Qwen3 MTP / Eagle3 drafts), 12 more KTO ↔ DPO alignment PRs (KTO promotion to stable is now in reach), three more {% generation %} chat templates (Gemma/Gemma 2, Phi-3, GLM-4-MoE), and a chunky SFT entropy bug fix.Full release notes: https://github.com/huggingface/trl/releases/tag/v1.3.0
qgallouedec
posted an update 6 months ago
Post
2890
TRL v1.2 introduces the SSDTrainer 🚀
Simple Self-Distillation (SSD) from Apple's paper "Embarrassingly Simple Self-Distillation Improves Code Generation" is now available as an experimental trainer in TRL.
The recipe is as minimal as the name suggests: sample completions from the model itself at a training-time temperature, then fine-tune on those raw, unverified samples with plain cross-entropy. No reward model. No verifier. No teacher model. No reinforcement learning. Just prompts and the model.
v1.2 also ships expanded tool-calling support (LLaMA 3.1 / 3.2, DeepSeek-V3), another round of KTO ↔ DPO alignment getting us closer to promoting KTO to stable, a big GRPO simplification for overlong tool results, deprecation of
Full release notes: https://github.com/huggingface/trl/releases/tag/v1.2.0
Simple Self-Distillation (SSD) from Apple's paper "Embarrassingly Simple Self-Distillation Improves Code Generation" is now available as an experimental trainer in TRL.
The recipe is as minimal as the name suggests: sample completions from the model itself at a training-time temperature, then fine-tune on those raw, unverified samples with plain cross-entropy. No reward model. No verifier. No teacher model. No reinforcement learning. Just prompts and the model.
from trl.experimental.ssd import SSDConfig, SSDTrainer
trainer = SSDTrainer(
model="Qwen/Qwen3-4B-Instruct",
args=SSDConfig(temperature=0.6, top_k=20, top_p=0.95),
train_dataset=dataset,
)
trainer.train()v1.2 also ships expanded tool-calling support (LLaMA 3.1 / 3.2, DeepSeek-V3), another round of KTO ↔ DPO alignment getting us closer to promoting KTO to stable, a big GRPO simplification for overlong tool results, deprecation of
use_transformers_paged, and key fixes for VLM response parsing.Full release notes: https://github.com/huggingface/trl/releases/tag/v1.2.0
qgallouedec
posted an update 6 months ago
Post
3150
TRL v1.0 is out!
Hugging Face's TRL library is downloaded 3 million times a month. Over 130k models trained with it are public on the Hub, and major projects like @unsloth and @axolotl-ai-co build directly on top of it. v1.0 is the moment we acknowledged that responsibility explicitly, with a real stability contract.
The field hasn't settled. Building stable software in a domain that keeps invalidating its own assumptions is the actual problem we're solving. The answer is a design that can absorb the next shift without breaking what people rely on.
What's in v1.0:
Deep Hugging Face integration, low infrastructure burden
What's next: asynchronous GRPO, better scaling support, and making training legible enough that agents can inspect and steer it.
Read more: hf.co/blog/trl-v1
Hugging Face's TRL library is downloaded 3 million times a month. Over 130k models trained with it are public on the Hub, and major projects like @unsloth and @axolotl-ai-co build directly on top of it. v1.0 is the moment we acknowledged that responsibility explicitly, with a real stability contract.
The field hasn't settled. Building stable software in a domain that keeps invalidating its own assumptions is the actual problem we're solving. The answer is a design that can absorb the next shift without breaking what people rely on.
What's in v1.0:
Deep Hugging Face integration, low infrastructure burden
What's next: asynchronous GRPO, better scaling support, and making training legible enough that agents can inspect and steer it.
pip install --upgrade trlRead more: hf.co/blog/trl-v1