Alexey Dontsov

therem's activity

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

upvoted a paper 11 days ago

Don't Blind Your VLA: Aligning Visual Representations for OOD Generalization

Paper • 2510.25616 • Published 19 days ago • 90

upvoted a paper 26 days ago

GAS: Improving Discretization of Diffusion ODEs via Generalized Adversarial Solver

Paper • 2510.17699 • Published 28 days ago • 24

upvoted a paper 27 days ago

Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs

Paper • 2510.11288 • Published Oct 13 • 47

upvoted 3 papers about 1 month ago

When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA

Paper • 2510.04849 • Published Oct 6 • 112

OrtSAE: Orthogonal Sparse Autoencoders Uncover Atomic Features

Paper • 2509.22033 • Published Sep 26 • 18

The Rogue Scalpel: Activation Steering Compromises LLM Safety

Paper • 2509.22067 • Published Sep 26 • 27

upvoted 2 papers 3 months ago

When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness Methods for LLMs

Paper • 2508.11383 • Published Aug 15 • 40

SONAR-LLM: Autoregressive Transformer that Thinks in Sentence Embeddings and Speaks in Tokens

Paper • 2508.05305 • Published Aug 7 • 46

upvoted 8 papers 5 months ago

Heeding the Inner Voice: Aligning ControlNet Training via Intermediate Features Feedback

Paper • 2507.02321 • Published Jul 3 • 39

Listener-Rewarded Thinking in VLMs for Image Preferences

Paper • 2506.22832 • Published Jun 28 • 23

Inverse-and-Edit: Effective and Fast Image Editing by Cycle Consistency Models

Paper • 2506.19103 • Published Jun 23 • 42

DreamBoothDPO: Improving Personalized Generation using Direct Preference Optimization

Paper • 2505.20975 • Published May 27 • 36

Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models

Paper • 2506.06395 • Published Jun 5 • 132

Geopolitical biases in LLMs: what are the "good" and the "bad" countries according to contemporary language models

Paper • 2506.06751 • Published Jun 7 • 71

Will It Still Be True Tomorrow? Multilingual Evergreen Question Classification to Improve Trustworthy QA

Paper • 2505.21115 • Published May 27 • 139

AmbiK: Dataset of Ambiguous Tasks in Kitchen Environment

Paper • 2506.04089 • Published Jun 4 • 47

upvoted an article 7 months ago

view article

Article

Process Reinforcement through Implicit Rewards

• 

Jan 3

•

31

upvoted a collection 7 months ago

InternLM2-Math

Collection

16 items • Updated Feb 11 • 9

upvoted a paper 7 months ago

TTRL: Test-Time Reinforcement Learning

Paper • 2504.16084 • Published Apr 22 • 120

upvoted a paper 8 months ago

Learn Your Reference Model for Real Good Alignment

Paper • 2404.09656 • Published Apr 15, 2024 • 89