The Curse and Blessing of Mean Bias in FP4-Quantized LLM Training
Paper
• 2603.10444 • Published
• 5
AI & ML interests
None defined yet.
Recent Activity
View all activity
xianbao
submitted a paper to Daily Papers about 23 hours ago
albertvillanova
posted an
update 16 days ago
Post
1885
🚀 TRL v0.29.0 introduces trl-training: an agent-native training skill.
This makes the TRL CLI a structured, agent-readable capability, allowing AI agents to reliably execute training workflows such as:
- Supervised Fine-Tuning (SFT)
- Direct Preference Optimization (DPO)
- Group Relative Policy Optimization (GRPO)
We’re excited to see what the community builds on top of this.
If you’re working on AI agents, alignment research, or scalable RL training infrastructure: give TRL v0.29.0 a try! 🤗
The future of ML tooling is agent-native.
🔗 https://github.com/huggingface/trl/releases/tag/v0.29.0
This makes the TRL CLI a structured, agent-readable capability, allowing AI agents to reliably execute training workflows such as:
- Supervised Fine-Tuning (SFT)
- Direct Preference Optimization (DPO)
- Group Relative Policy Optimization (GRPO)
We’re excited to see what the community builds on top of this.
If you’re working on AI agents, alignment research, or scalable RL training infrastructure: give TRL v0.29.0 a try! 🤗
The future of ML tooling is agent-native.
🔗 https://github.com/huggingface/trl/releases/tag/v0.29.0
albertvillanova
posted an
update about 1 month ago
Post
1754
5 years already working in democratizing AI 🤗
Grateful to be part of such an awesome team making it happen every day.
Grateful to be part of such an awesome team making it happen every day.
csukuangfj
authored 11 papers 4 months ago
Zipformer: A faster and better encoder for automatic speech recognition
Paper
• 2310.11230 • Published
• 1
PromptASR for contextualized ASR with controllable style
Paper
• 2309.07414 • Published
Libriheavy: a 50,000 hours ASR corpus with punctuation casing and context
Paper
• 2309.08105 • Published
Pruned RNN-T for fast, memory-efficient ASR training
Paper
• 2206.13236 • Published
Predicting Multi-Codebook Vector Quantization Indexes for Knowledge Distillation
Paper
• 2211.00508 • Published
Blank-regularized CTC for Frame Skipping in Neural Transducer
Paper
• 2305.11558 • Published
LibriheavyMix: A 20,000-Hour Dataset for Single-Channel Reverberant Multi-Talker Speech Separation, ASR and Speaker Diarization
Paper
• 2409.00819 • Published
Delay-penalized CTC implemented based on Finite State Transducer
Paper
• 2305.11539 • Published
k2SSL: A Faster and Better Framework for Self-Supervised Speech Representation Learning
Paper
• 2411.17100 • Published
ZipVoice: Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching
Paper
• 2506.13053 • Published
ZipVoice-Dialog: Non-Autoregressive Spoken Dialogue Generation with Flow Matching
Paper
• 2507.09318 • Published
• 2
dawood
authored a paper 4 months ago
xianbao
authored a paper 4 months ago
yuna0x0
authored a paper 6 months ago
osanseviero
authored a paper 6 months ago
albertvillanova
posted an
update 7 months ago
Post
4603
Latest smolagents release supports GPT-5: build agents that think, plan, and act.
⚡ Upgrade now and put GPT-5 to work!
⚡ Upgrade now and put GPT-5 to work!
- 1 reply
albertvillanova
posted an
update 7 months ago
Post
702
🚀 smolagents v1.21.0 is here!
Now with improved safety in the local Python executor: dunder calls are blocked!
⚠️ Still, not fully isolated: for untrusted code, use a remote executor instead: Docker, E2B, Wasm.
✨ Many bug fixes: more reliable code.
👉 https://github.com/huggingface/smolagents/releases/tag/v1.21.0
Now with improved safety in the local Python executor: dunder calls are blocked!
⚠️ Still, not fully isolated: for untrusted code, use a remote executor instead: Docker, E2B, Wasm.
✨ Many bug fixes: more reliable code.
👉 https://github.com/huggingface/smolagents/releases/tag/v1.21.0