Peter Szemraj's picture

Peter Szemraj PRO

pszemraj

·

https://pszemraj.carrd.co/

pszemraj

AI & ML interests

metallic intuition

Recent Activity

upvoted a paper 1 day ago

Survey on Evaluation of LLM-based Agents

upvoted a paper 1 day ago

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

upvoted a paper 1 day ago

Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models

View all activity

Organizations

pszemraj's activity

upvoted 3 papers 1 day ago

Survey on Evaluation of LLM-based Agents

Paper • 2503.16416 • Published 3 days ago • 60

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

Paper • 2503.13358 • Published 6 days ago • 82

Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models

Paper • 2503.16419 • Published 3 days ago • 53

upvoted 2 papers 3 days ago

Optimizing Decomposition for Optimal Claim Verification

Paper • 2503.15354 • Published 4 days ago • 18

φ-Decoding: Adaptive Foresight Sampling for Balanced Inference-Time Exploration and Exploitation

Paper • 2503.13288 • Published 6 days ago • 46

upvoted a paper 6 days ago

Transformers without Normalization

Paper • 2503.10622 • Published 10 days ago • 130

upvoted a paper 11 days ago

Implicit Reasoning in Transformers is Reasoning through Shortcuts

Paper • 2503.07604 • Published 13 days ago • 21

upvoted 4 papers 12 days ago

WritingBench: A Comprehensive Benchmark for Generative Writing

Paper • 2503.05244 • Published 16 days ago • 17

DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs

Paper • 2503.07067 • Published 13 days ago • 29

SEAP: Training-free Sparse Expert Activation Pruning Unlock the Brainpower of Large Language Models

Paper • 2503.07605 • Published 13 days ago • 65

Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders

Paper • 2503.03601 • Published 18 days ago • 215

upvoted 2 papers 13 days ago

HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization

Paper • 2503.04598 • Published 17 days ago • 18

EuroBERT: Scaling Multilingual Encoders for European Languages

Paper • 2503.05500 • Published 16 days ago • 75

upvoted 2 papers 14 days ago

L^2M: Mutual Information Scaling Law for Long-Context Language Modeling

Paper • 2503.04725 • Published 17 days ago • 19

LLM as a Broken Telephone: Iterative Generation Distorts Information

Paper • 2502.20258 • Published 24 days ago • 24

upvoted a paper 15 days ago

RedPajama: an Open Dataset for Training Large Language Models

Paper • 2411.12372 • Published Nov 19, 2024 • 53

upvoted a paper 17 days ago

HoT: Highlighted Chain of Thought for Referencing Supporting Facts from Inputs

Paper • 2503.02003 • Published 20 days ago • 44

upvoted 2 papers 19 days ago

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

Paper • 2503.01688 • Published 20 days ago • 19

From Hours to Minutes: Lossless Acceleration of Ultra Long Sequence Generation up to 100K Tokens

Paper • 2502.18890 • Published 25 days ago • 27