works on

From the 1 of 21 linked papers with an AI index.

activity
20242026
collaborators
Showing cs.LGShow all

10 papers · 1 filter

cs.LG2026

Leveraging Instruction Tuning and Merging for Reasoning Model Adaptation

Yu-Du Feng, Niels Mündler-Sasahara, Mark Vero +1

The paper proposes a cheap method to adapt reasoning language models to new tasks by first instruction‑tuning them on ordinary supervised data and then merging the tuned model back…

cs.LG2026

Widening the Gap: Exploiting LLM Quantization via Outlier Injection

Xiaohua Zhan, Kazuki Egashira, Robin Staab +2

LLM quantization has become essential for memory-efficient deployment. Recent work has shown that quantization schemes can pose critical security risks: an adversary may release a…

cs.LG2026

Delay, Plateau, or Collapse: Evaluating the Impact of Systematic Verification Error on RLVR

Kazuki Egashira, Mark Vero, Jasper Dekoninck +3

Reinforcement Learning with Verifiable Rewards (RLVR) has become a powerful approach for improving the reasoning capabilities of large language models (LLMs). While RLVR is designe…

cs.LG2026

Fewer Weights, More Problems: A Practical Attack on LLM Pruning

Kazuki Egashira, Robin Staab, Thibaud Gloaguen +2

Model pruning, i.e., removing a subset of model weights, has become a prominent approach to reducing the memory footprint of large language models (LLMs) during inference. Notably,…

cs.LG2025

Pay Attention to the Triggers: Constructing Backdoors That Survive Distillation

Giovanni De Muri, Mark Vero, Robin Staab +1

LLMs are often used by downstream users as teacher models for knowledge distillation, compressing their capabilities into memory-efficient models. However, as these teacher models…

cs.LG2025

Watch your steps: Dormant Adversarial Behaviors that Activate upon LLM Finetuning

Thibaud Gloaguen, Mark Vero, Robin Staab +1

Finetuning open-weight Large Language Models (LLMs) is standard practice for achieving task-specific performance improvements. Until now, finetuning has been regarded as a controll…