7 30 12

Nikita Sushko

chameleon-lizard

http://chameleon-lizard.ru:81

chameleon-lizard

AI & ML interests

NLP, Multilingual Models, Multiagent Systems

Recent Activity

new activity about 23 hours ago

featherless-ai/Qwerky-QwQ-32B:Is the source code for this conversion available?

upvoted a paper 9 days ago

I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders

upvoted a paper 10 days ago

When Less is Enough: Adaptive Token Reduction for Efficient Image Representation

View all activity

Organizations

chameleon-lizard's activity

upvoted a paper 9 days ago

I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders

Paper • 2503.18878 • Published 10 days ago • 110

upvoted a paper 10 days ago

When Less is Enough: Adaptive Token Reduction for Efficient Image Representation

Paper • 2503.16660 • Published 13 days ago • 70

upvoted a paper 13 days ago

One-Step Residual Shifting Diffusion for Image Super-Resolution via Distillation

Paper • 2503.13358 • Published 17 days ago • 90

upvoted a paper 15 days ago

RWKV-7 "Goose" with Expressive Dynamic State Evolution

Paper • 2503.14456 • Published 16 days ago • 130

upvoted 2 papers 24 days ago

EuroBERT: Scaling Multilingual Encoders for European Languages

Paper • 2503.05500 • Published 27 days ago • 74

RuCCoD: Towards Automated ICD Coding in Russian

Paper • 2502.21263 • Published Feb 28 • 126

upvoted a collection 26 days ago

SynthDetoxM

Collection

Data and models from NAACL 2025 paper "SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators" by Moskovskiy et al. • 4 items • Updated 28 days ago • 2

upvoted a paper 30 days ago

When an LLM is apprehensive about its answers -- and when its uncertainty is justified

Paper • 2503.01688 • Published about 1 month ago • 20

upvoted 4 papers about 1 month ago

GHOST 2.0: generative high-fidelity one shot transfer of heads

Paper • 2502.18417 • Published Feb 25 • 65

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

Paper • 2502.15007 • Published Feb 20 • 169

How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?

Paper • 2502.14502 • Published Feb 20 • 88

Cramming 1568 Tokens into a Single Vector and Back Again: Exploring the Limits of Embedding Space Capacity

Paper • 2502.13063 • Published Feb 18 • 68

upvoted 4 papers about 2 months ago

upvoted 4 papers 2 months ago

Exploring Methods for Cross-lingual Text Style Transfer: The Case of Text Detoxification

Paper • 2311.13937 • Published Nov 23, 2023 • 1

MERA: A Comprehensive LLM Evaluation in Russian

Paper • 2401.04531 • Published Jan 9, 2024 • 2

Demarked: A Strategy for Enhanced Abusive Speech Moderation through Counterspeech, Detoxification, and Message Management

Paper • 2406.19543 • Published Jun 27, 2024 • 1

Multilingual and Explainable Text Detoxification with Parallel Corpora

Paper • 2412.11691 • Published Dec 16, 2024 • 1