Peter Szemraj PRO

pszemraj

https://pszemraj.carrd.co/

AI & ML interests

metallic intuition

Recent Activity

upvoted a paper 1 day ago

Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning

upvoted a paper 1 day ago

LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Training

upvoted a paper 1 day ago

LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis

View all activity

Organizations

upvoted 3 papers 1 day ago

Skill0.5: Joint Skill Internalization and Utilization for Out-of-Distribution Generalization in Agentic Reinforcement Learning

Paper • 2605.28424 • Published 7 days ago • 28

LaRA: Layer-wise Representation Analysis for Detecting Data Contamination in RL Post-Training

Paper • 2605.29888 • Published 6 days ago • 27

LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis

Paper • 2605.30434 • Published 6 days ago • 17

upvoted a paper 9 days ago

Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

Paper • 2605.22791 • Published 13 days ago • 30

upvoted an article 12 days ago

Article

Introducing the Ettin Reranker Family

tomaarsen

•

15 days ago

• 50

upvoted a paper 13 days ago

Hierarchical Reasoning Model

Paper • 2506.21734 • Published Jun 26, 2025 • 54

upvoted 3 papers 20 days ago

upvoted an article 21 days ago

Article

Adding Benchmaxxer Repellant to the Open ASR Leaderboard

bezzam, Steveeeeeeen, eustlb, SBruccoleriAppen, jmss-appen, c-e-ford-appen, wgb14, YukaiHuang, like2026, logicbean, ally-lxl

•

28 days ago

• 17

upvoted a paper 21 days ago

Investigating Efficiently Extending Transformers for Long Input Summarization

Paper • 2208.04347 • Published Aug 8, 2022 • 1

upvoted an article 22 days ago

Article

EMO: Pretraining mixture of experts for emergent modularity

allenai

•

25 days ago

• 38

upvoted an article 29 days ago

Article

Multimodal Embedding & Reranker Models with Sentence Transformers

tomaarsen

•

Apr 9

• 61

upvoted a collection about 1 month ago

OlmPool

Collection

Collection of models from the paper "Cracks in the Foundation: Seemingly Minor Architectural Choices Impact Long Context Extension". • 26 items • Updated Apr 30 • 5

upvoted 3 papers about 1 month ago

A Survey on LLM-based Conversational User Simulation

Paper • 2604.24977 • Published Apr 27 • 8

Efficient Training on Multiple Consumer GPUs with RoundPipe

Paper • 2604.27085 • Published Apr 29 • 46

Why Fine-Tuning Encourages Hallucinations and How to Fix It

Paper • 2604.15574 • Published Apr 16 • 25

upvoted a collection about 1 month ago

Olmo 3.1

Collection

The latest members of the Olmo 3 family: another 3 weeks of RL for 32B Think, the 32B Instruct model, large post-training research datasets... • 9 items • Updated Dec 23, 2025 • 52

upvoted an article about 1 month ago

Article

Granite 4.1 LLMs: How They’re Built

ibm-granite

•

Apr 29

• 77

upvoted a paper about 1 month ago

Programming with Data: Test-Driven Data Engineering for Self-Improving LLMs from Raw Corpora

Paper • 2604.24819 • Published Apr 27 • 89

Peter Szemraj PRO

AI & ML interests

Recent Activity

Organizations

pszemraj's activity

Introducing the Ettin Reranker Family

Adding Benchmaxxer Repellant to the Open ASR Leaderboard

EMO: Pretraining mixture of experts for emergent modularity

Multimodal Embedding & Reranker Models with Sentence Transformers

Granite 4.1 LLMs: How They’re Built