J C

dark-pen's activity

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

liked a model about 16 hours ago

McGill-NLP/A3-Qwen3.5-4B

Image-Text-to-Text • 5B • Updated Apr 16 • 7 • 2

liked a dataset about 16 hours ago

McGill-NLP/A3-Synth

Updated Apr 10 • 3.29k • 1

upvoted a paper about 16 hours ago

Structured Distillation of Web Agent Capabilities Enables Generalization

Paper • 2604.07776 • Published Apr 9 • 23

upvoted a collection about 16 hours ago

A3: Agent-as-Annotators

Collection

Models and data from "Structured Distillation of Web Agent Capabilities Enables Generalization" (arXiv:2604.07776) • 6 items • Updated Apr 14 • 1

liked a Space about 17 hours ago

Running

Agents

Featured

15

PTS Visualizer

🔍

15

Visualize pivotal tokens and thought anchors in language models

liked a dataset 6 days ago

PsiBotAI/SynData

Viewer • Updated 4 days ago • 449k • 173k • 174

upvoted a paper 6 days ago

Lance: Unified Multimodal Modeling by Multi-Task Synergy

Paper • 2605.18678 • Published 8 days ago • 72

liked 2 models 6 days ago

bytedance-research/Lance

Any-to-Any • Updated 3 days ago • 1.68k • 815

SulphurAI/Sulphur-2-base

Text-to-Video • 9B • Updated 4 days ago • 1.35M • 1.35k

upvoted a paper 13 days ago

Fast and Effective On-policy Distillation from Reasoning Prefixes

Paper • 2602.15260 • Published Feb 16 • 1

liked a dataset 13 days ago

lllyx/OpenThought3-Qwen3-4B

Viewer • Updated 14 days ago • 305k • 149 • 2

upvoted a paper 13 days ago

Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe

Paper • 2604.13016 • Published Apr 14 • 107

liked 3 models 14 days ago

TomoroAI/tomoro-colqwen3-embed-4b

Visual Document Retrieval • 4B • Updated Dec 7, 2025 • 196k • 27

leo-vnuuet/ColQwen3.5-2B-Embedding

Feature Extraction • Updated Mar 29 • 19 • 2

leo-vnuuet/ColQwen3.5-0.8B-Embedding

Feature Extraction • Updated Mar 29 • 17 • 6

upvoted 4 collections 14 days ago

Papers

Collection

3 items • Updated Mar 31 • 1

ColGemma4 — Gemma-4 Visual Retrieval

Collection

ColBERT-style late-interaction visual document retrieval adapters built on Google Gemma-4 (E2B and E4B variants). • 2 items • Updated Apr 18 • 1

ColQwen3.5 — Qwen3.5 Visual Retrieval

Collection

Visual document retrieval models on Qwen3.5 backbone. ViDoRe v3 leaderboard competitors, 128-dim multi-vector. • 2 items • Updated Apr 13 • 1

Hydra — Dual-Head Retrieval and Generation

Collection

Dual-head VLM: ColBERT retrieval + autoregressive generation by toggling one LoRA. Canonical 4B + 0.8B, omni proof-of-concept, baselines. • 4 items • Updated Apr 19 • 1

liked a model 14 days ago

athrael-soju/DualHead-GritLM-Qwen3.5-4B

Visual Document Retrieval • Updated Apr 18 • 11 • 1