Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation
Paper β’ 2604.27263 β’ Published β’ 8
AI & ML interests
π€ Hugging Face x πΈ BigScience initiative to create open source community resources for LAMs.
Recent Activity
View all activity
gigantΒ
authored a
paper 2 days ago
gigantΒ
submitted a
paper to Daily Papers 3 days ago
[bot] Conversion to Parquet
#1 opened 4 days ago
by 
parquet-converter
davanstrienΒ
updated a
dataset 3 days ago
gigantΒ
authored a
paper 10 days ago
stefan-itΒ
submitted a
paper to Daily Papers 12 days ago
Add dataset card body: description, subsets, model link, citation, limitations
#2 opened 19 days ago
by 
davanstrien
stefan-itΒ
submitted a
paper to Daily Papers 20 days ago
davanstrienΒ
updated a
dataset 24 days ago
[bot] Conversion to Parquet
#2 opened over 2 years ago
by parquet-converter
Add dataset card
#3 opened 24 days ago
by davanstrien
[bot] Conversion to Parquet
#1 opened about 2 months ago
by parquet-converter
davanstrienΒ
updated a
dataset about 2 months ago
davanstrienΒ
published a
dataset about 2 months ago
albertvillanovaΒ
posted an update 3 months ago
Post
2754
π 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)
Weβre excited to see what the community builds on top of this.
If youβre working on AI agents, alignment research, or scalable RL training infrastructure: give TRL v0.29.0 a try! π€
The future of ML tooling is agent-native.
π 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)
Weβre excited to see what the community builds on top of this.
If youβre working on AI agents, alignment research, or scalable RL training infrastructure: give TRL v0.29.0 a try! π€
The future of ML tooling is agent-native.
π https://github.com/huggingface/trl/releases/tag/v0.29.0
albertvillanovaΒ
posted an update 3 months ago
Post
1971
5 years already working in democratizing AI π€
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.
stefan-itΒ
submitted a
paper to Daily Papers 4 months ago
christopherΒ
authored a
paper 6 months ago
ZaidΒ
authored a
paper 7 months ago
stefan-itΒ
authored a
paper 7 months ago