Lukas Galke Poech

Lukas Galke Poech on Hugging Face Daily Papers: 10 papers, 0 in the top 3 of their day, 106 upvotes.

1. Selecting The Most Informative Tokens in Natural Language Autoencoders 16 upvotes, #49 of 2026-09-30
2. DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data 32 upvotes, #8 of 2026-08-17
3. The Arbiter Agent: Continually Monitoring Multi-Agent Conversations to Detect Emergent Misalignment 13 upvotes, #19 of 2026-06-15
4. BrainSurgery: Reproducible and Reliable Declarative Weight Manipulations for Model Editing and Upcycling 8 upvotes, #22 of 2026-06-10
5. PsychoSafe: Eliciting Psychologically-Informed Refusals in Large Language Models 7 upvotes, #23 of 2026-06-10
6. LLMs Can Leak Training Data But Do They Want To? A Propensity-Aware Evaluation of Memorization in LLMs 8 upvotes, #19 of 2026-06-05
7. Emergent Languages in Populations of Language Model Agents: From Token Efficiency to Oversight Evasion 11 upvotes, #29 of 2026-06-01
8. Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals 12 upvotes, #24 of 2026-05-27
9. FlexMoRE: A Flexible Mixture of Rank-heterogeneous Experts for Efficient Federatedly-trained Large Language Models 2 upvotes, #42 of 2026-02-10
10. Dynaword: From One-shot to Continuously Developed Datasets 13 upvotes, #9 of 2025-08-05

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.