Open to Collab

Solomatin Roman

Samoed

AI & ML interests

None yet

Recent Activity

reacted to HannesVonEssen's post with ❤️ 2 days ago

📣 I made a visualizer for Hugging Face models: https://hfviewer.com ✨ Simply paste a Hugging Face URL to get an interactive visualization of the architecture! 🔗 The recent Qwen3.6-27B model as an example: https://hfviewer.com/Qwen/Qwen3.6-27B Feel free to try it out and give me feedback on how it can be improved! ❤️

updated a dataset 4 days ago

DeepPavlov/WebLINX-ru

upvoted a paper 5 days ago

GradMem: Learning to Write Context into Memory with Test-Time Gradient Descent

View all activity

Organizations

upvoted a paper 5 days ago

GradMem: Learning to Write Context into Memory with Test-Time Gradient Descent

Paper • 2603.13875 • Published Mar 14 • 36

upvoted a paper 9 days ago

Prism-Reranker: Beyond Relevance Scoring -- Jointly Producing Contributions and Evidence for Agentic Retrieval

Paper • 2604.23734 • Published 11 days ago • 3

upvoted an article 15 days ago

Article

DenseOn with the LateOn: Open State-of-the-Art Single and Multi-Vector Models

16 days ago

•

upvoted an article 20 days ago

Article

Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers

21 days ago

•

upvoted an article 28 days ago

Article

Multimodal Embedding & Reranker Models with Sentence Transformers

28 days ago

•

upvoted a paper 28 days ago

Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation for Dense Retrieval

Paper • 2604.04734 • Published Apr 6 • 12

upvoted an article 28 days ago

Article

BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders

29 days ago

•

upvoted a paper 28 days ago

BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs

Paper • 2604.02045 • Published Apr 2 • 35

upvoted a paper about 2 months ago

LMEB: Long-horizon Memory Embedding Benchmark

Paper • 2603.12572 • Published Mar 13 • 73

upvoted an article 3 months ago

Article

ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models?

Feb 19

•

upvoted 2 papers 3 months ago

ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models

Paper • 2602.16609 • Published Feb 18 • 7

MAEB: Massive Audio Embedding Benchmark

Paper • 2602.16008 • Published Feb 17 • 25

upvoted a collection 3 months ago

LateOn-Code 💻

Collection

State-of-the-art late interaction code retrieval models • 6 items • Updated 30 days ago • 19

upvoted 3 articles 3 months ago

Article

LateOn-Code & ColGrep: LightOn unveils state-of-the-art code retrieval models and code search tooling

Feb 12

•

Article

Community Evals: Because we're done trusting black-box leaderboards over the community

Feb 4

•

Article

Nemotron ColEmbed V2: Raising the Bar for Multimodal Retrieval with ViDoRe V3’s Top Model

Feb 4

•

upvoted an article 4 months ago

Article

🥃 Distilling Tiny Embeddings

Jan 10

•

upvoted a collection 4 months ago

Qwen3-VL-Embedding

Collection

2 items • Updated Jan 8 • 67

upvoted a collection 5 months ago

SauerkrautLM-Vision-Document-Retrieval

Collection

7 items • Updated Dec 15, 2025 • 9

upvoted a paper 5 months ago

T-pro 2.0: An Efficient Russian Hybrid-Reasoning Model and Playground

Paper • 2512.10430 • Published Dec 11, 2025 • 119

Solomatin Roman

AI & ML interests

Recent Activity

Organizations

Samoed's activity

DenseOn with the LateOn: Open State-of-the-Art Single and Multi-Vector Models

Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers

Multimodal Embedding & Reranker Models with Sentence Transformers

BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders

**ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models?**

LateOn-Code & ColGrep: LightOn unveils state-of-the-art code retrieval models and code search tooling

Community Evals: Because we're done trusting black-box leaderboards over the community

Nemotron ColEmbed V2: Raising the Bar for Multimodal Retrieval with ViDoRe V3’s Top Model

🥃 Distilling Tiny Embeddings

ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models?