collaborators

7 papers

cs.AI2026

OpenThoughts-Agent: Data Recipes for Agentic Models

Negin Raoof, Richard Zhuang, Marianna Nezhurina +47

Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts…

cs.LG2026

Trace Reconstruction with Language Models

Franziska Weindel, Michael Girsch, Reinhard Heckel

The general trace reconstruction problem seeks to recover an original sequence from its noisy copies independently corrupted by insertions, deletions, and substitutions. This probl…

cs.AI2026

LLM-Guided Search for Deletion-Correcting Codes

Franziska Weindel, Reinhard Heckel

Finding deletion-correcting codes of maximum size has been an open problem for over 70 years, even for a single deletion. We adapt FunSearch, a large language model (LLM)-guided ev…

cs.LG2026

Asymmetric Prompt Weighting for Reinforcement Learning with Verifiable Rewards

Reinhard Heckel, Mahdi Soltanolkotabi, Christos Thramboulidis

Reinforcement learning with verifiable rewards has driven recent advances in LLM post-training, in particular for reasoning. Policy optimization algorithms generate a number of res…

cs.IT2026

Near Optimal Code Construction for the Adversarial Torn Paper Channel With Edit Errors

Maria Abu-Sini, Reinhard Heckel

Motivated by DNA storage systems and 3D fingerprinting, this work studies the adversarial torn paper channel with edit errors. This channel first applies at most edit errors…

cs.IT2025

Transformer-Based Decoding in Concatenated Coding Schemes Under Synchronization Errors

Julian Streit, Franziska Weindel, Reinhard Heckel

We consider the reconstruction of a codeword from multiple noisy copies that are independently corrupted by insertions, deletions, and substitutions. This problem arises, for examp…