-
Offline Reinforcement Learning for LLM Multi-Step Reasoning
Paper • 2412.16145 • Published • 38 -
SPLADE-v3: New baselines for SPLADE
Paper • 2403.06789 • Published • 5 -
MMTEB: Massive Multilingual Text Embedding Benchmark
Paper • 2502.13595 • Published • 44 -
Reducing the Footprint of Multi-Vector Retrieval with Minimal Performance Impact via Token Pooling
Paper • 2409.14683 • Published • 11
Abdulhakeem Adefioye
kokolamba
AI & ML interests
None yet
Recent Activity
updated
a dataset
24 days ago
kokolamba/keen_popqa_gpt2xl_generations
upvoted
a
paper
about 1 month ago
Estimating Knowledge in Large Language Models Without Generating a
Single Token
published
a dataset
about 2 months ago
kokolamba/keen_popqa_gpt2xl_generations
Organizations
-
Offline Reinforcement Learning for LLM Multi-Step Reasoning
Paper • 2412.16145 • Published • 38 -
SPLADE-v3: New baselines for SPLADE
Paper • 2403.06789 • Published • 5 -
MMTEB: Massive Multilingual Text Embedding Benchmark
Paper • 2502.13595 • Published • 44 -
Reducing the Footprint of Multi-Vector Retrieval with Minimal Performance Impact via Token Pooling
Paper • 2409.14683 • Published • 11
models
29
kokolamba/moe-mha
Updated
kokolamba/moe-kv-128
Updated
kokolamba/moe-o-192
Updated
•
3
kokolamba/SubspaceDecoder_mha
Updated
kokolamba/SubspaceDecoder_mla192-0-0
Updated
kokolamba/SubspaceDecoder_mla0-0-0
Updated
kokolamba/SubspaceDecoder_mla0-0-192
Updated
kokolamba/SubspaceDecoder_mla0-96-192
Updated
kokolamba/SubspaceDecoder_mla0-128-0
Updated
kokolamba/SubspaceDecoder_mla192-96-192
Updated