Article
24
Nicolas-BZRD
Nicolas-BZRD
AI & ML interests
PhD Student | NLP - LLMs - Adaptation real-world problem
Optimization
Recent Activity
liked a model about 1 hour ago
BidirLM/BidirLM-Omni-2.5B-Embedding updated a collection about 18 hours ago
BidirLM updated a collection about 18 hours ago
BidirLMOrganizations
Articles 4
Article
12
When Does Reasoning Matter? Unpacking the Contribution of Reasoning to LLM Performance
BidirLM is a family of 5 frontier bidirectional encoders, including an omnimodal variant at 2.5B.
-
BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs
Paper • 2604.02045 • Published • 29 -
BidirLM/BidirLM-Omni-2.5B-Embedding
Sentence Similarity • 2B • Updated • 2.68k • 7 -
BidirLM/BidirLM-1.7B-Embedding
Sentence Similarity • 2B • Updated • 23 • 2 -
BidirLM/BidirLM-1B-Embedding
Sentence Similarity • 1.0B • Updated • 194 • 1
A Controlled Study of Reasoning's Contribution to Model Performance.
-
When Does Reasoning Matter? A Controlled Study of Reasoning's Contribution to Model Performance
Paper • 2509.22193 • Published • 38 -
When-Does-Reasoning-Matter/general-reasoning-ift-pairs
Viewer • Updated • 2.97M • 198 • 4 -
When-Does-Reasoning-Matter/math-reasoning-ift-pairs
Viewer • Updated • 458k • 136 • 8
BidirLM is a family of 5 frontier bidirectional encoders, including an omnimodal variant at 2.5B.
-
BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs
Paper • 2604.02045 • Published • 29 -
BidirLM/BidirLM-Omni-2.5B-Embedding
Sentence Similarity • 2B • Updated • 2.68k • 7 -
BidirLM/BidirLM-1.7B-Embedding
Sentence Similarity • 2B • Updated • 23 • 2 -
BidirLM/BidirLM-1B-Embedding
Sentence Similarity • 1.0B • Updated • 194 • 1
A Controlled Study of Reasoning's Contribution to Model Performance.
-
When Does Reasoning Matter? A Controlled Study of Reasoning's Contribution to Model Performance
Paper • 2509.22193 • Published • 38 -
When-Does-Reasoning-Matter/general-reasoning-ift-pairs
Viewer • Updated • 2.97M • 198 • 4 -
When-Does-Reasoning-Matter/math-reasoning-ift-pairs
Viewer • Updated • 458k • 136 • 8
models 92
Nicolas-BZRD/mt0-base_dialogsum_Mistral-7B-Instruct-v0.2_uld_loss
0.6B • Updated • 2
Nicolas-BZRD/mt0-base_dialogsum_Mistral-7B-Instruct-v0.2_text_teacher
0.6B • Updated • 4
Nicolas-BZRD/mt0-base_pubmed_qa_Llama-2-7b-chat-hf_uld_loss
0.6B • Updated • 3
Nicolas-BZRD/mt0-base_pubmed_qa_Llama-2-7b-chat-hf_text_teacher
0.6B • Updated • 1
Nicolas-BZRD/mt0-base_qed_Llama-2-7b-chat-hf_uld_loss
0.6B • Updated • 7
Nicolas-BZRD/mt0-base_qed_Llama-2-7b-chat-hf_text_teacher
0.6B • Updated • 2
Nicolas-BZRD/mt0-base_dialogsum_Llama-2-7b-chat-hf_uld_loss
0.6B • Updated • 4
Nicolas-BZRD/mt0-base_dialogsum_Llama-2-7b-chat-hf_text_teacher
0.6B • Updated • 1
Nicolas-BZRD/pythia-160m-deduped_FairytaleQA_Llama-2-7b-chat-hf_uld_loss
Text Generation • 0.2B • Updated • 5
Nicolas-BZRD/pythia-160m-deduped_FairytaleQA_Llama-2-7b-chat-hf_text_teacher
Text Generation • 0.2B • Updated • 6
datasets 37
Nicolas-BZRD/AIME_1983_2025
Viewer • Updated • 979 • 9
Nicolas-BZRD/NanoBigBench
Viewer • Updated • 1.14k • 9
Nicolas-BZRD/NanoEmotion
Viewer • Updated • 500 • 12
Nicolas-BZRD/NanoMMLU
Viewer • Updated • 500 • 4
Nicolas-BZRD/gsm8k-ar-Qwen2-72B-Instruct
Viewer • Updated • 7.47k • 10
Nicolas-BZRD/gsm8k-ar-Meta-Llama-3.1-70B-Instruct
Viewer • Updated • 7.47k • 8
Nicolas-BZRD/gsm8k-ar-gemma-2-27b-it
Viewer • Updated • 7.47k • 6
Nicolas-BZRD/uld_loss_Mistral-7B-Instruct-v0.2-pubmed_qa_50k
Viewer • Updated • 50.5k • 8
Nicolas-BZRD/uld_loss_Llama-2-7b-chat-hf-squad
Viewer • Updated • 87.6k • 5
Nicolas-BZRD/uld_loss_Mistral-7B-Instruct-v0.2-squad
Viewer • Updated • 87.6k • 6