Robin Williams

bfuzzy1's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

liked a Space 3 days ago

Running

25

LFM2.5 Edge Research Agent

🔬

25

Run an on-device LFM2.5 research agent with WebGPU

New activity in rl-llm-wiki/knowledge-base 22 days ago

dedup: give the Gao over-optimization law one owning article

2

#650 opened 22 days ago by

lvwerra

figures: first two wiki figures (overoptimization turnover + RLHF two-KLs pipeline)

2

#648 opened 22 days ago by

lvwerra

new node: foundations/learning-path (the entry ramp)

2

#649 opened 22 days ago by

lvwerra

New activity in rl-llm-wiki/knowledge-base 23 days ago

new node: foundations/controllable-generation (de-orphans 5; the steering paradigm RLHF descends from)

2

#640 opened 23 days ago by

lvwerra

New activity in rl-llm-wiki/knowledge-base 24 days ago

topic: cross-link entropy-and-exploration + reasoning-emergence -> reporting-gap-audit

2

#615 opened 24 days ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base 25 days ago

topic: hallucination-and-abstention developing->comprehensive (o1-card frontier weave + monitor-as-verifier; 16 anchors verified)

2

#596 opened 25 days ago by

kshitijthakkar

source: arxiv:2412.16339 — Deliberative Alignment (Reasoning Enables Safer LMs)

6

#595 opened 26 days ago by

thomwolf

upvoted 2 papers 26 days ago

Gemma 4 Technical Report

Paper • 2607.02770 • Published Jul 2 • 75

Hierarchical Sparse Attention Done Right: Toward Infinite Context Modeling

Paper • 2607.02980 • Published Jul 3 • 84

New activity in rl-llm-wiki/knowledge-base 28 days ago

topic: policy-gradient-methods — deepen + add citations

2

#594 opened 28 days ago by

bfuzzy1

topic: judging-bias-and-contamination — deepen + add citations

3

#593 opened 28 days ago by

bfuzzy1

source: arxiv:2005.01643 — Offline RL: Tutorial, Review, and Perspectives on Open Problems

2

#592 opened 28 days ago by

bfuzzy1

source: arxiv:1910.00177 — Advantage-Weighted Regression (AWR)

2

#591 opened 28 days ago by

bfuzzy1

source: arxiv:2006.04779 — Conservative Q-Learning for Offline Reinforcement Learning

3

#590 opened 28 days ago by

bfuzzy1

topic: preference-reward-models — deepen + bump to comprehensive

2

#589 opened 28 days ago by

bfuzzy1

source: arxiv:1707.01495 — Hindsight Experience Replay

2

#588 opened 28 days ago by

bfuzzy1

topic: kl-regularization — build out from stub

2

#587 opened 28 days ago by

bfuzzy1

source: arxiv:2004.07219 — D4RL: Datasets for Deep Data-Driven Reinforcement Learning

2

#586 opened 28 days ago by

bfuzzy1

source: arxiv:1609.05473 — SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient

3

#585 opened 28 days ago by

bfuzzy1