2 citations · 2 across the 14 of their papers we have counts for
18 papers
Different Tasks, Different Timesteps: Temporal Task-Aware Quantization for Diffusion Language Models
Ofek Avivi, Eliron Rahimi, Avi Mendelson +1
Diffusion Language Models (DLMs) introduce unique quantization challenges due to their iterative denoising process. In this work, we reveal that quantization sensitivity is both ta…
On the SPOT: Enhancing Speculative Decoding for Diffusion Language Models via Commit-Order Prediction
Eliezer Shlomi, Eliron Rahimi, Avi Mendelson +1
Diffusion language models (DLMs) promise parallel text generation, but in practice face an unresolved speed-quality trade-off. Exact decoding fills in only one masked position per…
Generalization Is Stability, Not Accuracy: Multi-Axis Evaluation of LLMs
Nagham Omar, Mahmoud Jabarin, Maya Rozenshtein +3
Generalization in large language models (LLMs) is the ability to produce consistent and semantically stable outputs when the same input is expressed in different ways. Existing wor…
AnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation Repulsion
Gal Wertheizer, Rom Himelstein, Tomer Peretz +1
Adversarial attacks optimized on a single open-weight LLM can transfer to and jailbreak architecturally different models, allowing an attacker with white-box access to one model to…
LGNNIC: Acceleration of Large-Scale GNN Training using SmartNICs
Liad Gerstman, Aditya Dhakal, Dejan Milojicic +1
Graph Neural Networks (GNNs) are widely used across domains such as natural sciences, social network analysis, chip design, and recommendation systems. However, as graph sizes grow…
Linguistic Firewall: Geometry as Defense in Multi-Agent Systems Routing
Dvir Alsheich, Adar Peleg, Ben Hagag +3
The rapid integration of Large Language Models (LLMs) has driven the evolution of Multi-Agent Systems (MAS), where specialized agents collaborate to execute complex workflows. Effe…
Breaking the Bubble: Asynchronous Pipeline Parallel Training with Bounded Weight Inconsistency
Itay Elam, Eliron Rahimi, Avi Mendelson +1
Pipeline parallelism is essential for training large neural networks, but existing schedules trade off throughput, memory, and optimization consistency. Synchronous pipelines prese…
Extracting Recurring Vulnerabilities from Black-Box LLM-Generated Software
Tomer Kordonsky, Amit LeVi, Maayan Yamin +2
LLMs are increasingly used for code generation, but their outputs often follow recurring templates that can induce predictable vulnerabilities. We study vulnerability persistence i…
Step-Wise Refusal Dynamics in Autoregressive and Diffusion Language Models
Eliron Rahimi, Elad Hirshel, Rom Himelstein +3
Diffusion language models (DLMs) have recently emerged as a competitive alternative to autoregressive (AR) models, offering parallel decoding, competitive generation quality, and i…
REMIND: Input Loss Landscapes Reveal Residual Memorization in Post-Unlearning LLMs
Liran Cohen, Yaniv Nemcovesky, Avi Mendelson
Machine unlearning aims to remove the influence of specific training data from a model without requiring full retraining. This capability is crucial for ensuring privacy, safety, a…
You Had One Job: Per-Task Quantization Using LLMs' Hidden Representations
Amit LeVi, Raz Lapid, Rom Himelstein +3
Many LLM applications require only narrow capabilities, yet standard post-training quantization (PTQ) methods allocate precision without considering the target task. This can waste…
Silenced Biases: The Dark Side LLMs Learned to Refuse
Rom Himelstein, Amit LeVi, Brit Youngmann +2
Safety-aligned large language models (LLMs) are becoming increasingly widespread, especially in sensitive applications where fairness is essential and biased outputs can cause sign…
Silent Tokens, Loud Effects: Padding in LLMs
Rom Himelstein, Amit LeVi, Yonatan Belinkov +1
Padding tokens are widely used in large language models (LLMs) to equalize sequence lengths during batched inference. While they should be fully masked, implementation errors can c…
Representing LLMs in Prompt Semantic Task Space
Idan Kashani, Avi Mendelson, Yaniv Nemcovsky
Large language models (LLMs) achieve impressive results over various tasks, and ever-expanding public repositories contain an abundance of pre-trained models. Therefore, identifyin…
: Reconstruction, Raw, and Rain: Deraining Directly in the Bayer Domain
Nate Rothschild, Moshe Kimhi, Avi Mendelson +1
Image reconstruction from corrupted images is crucial across many domains. Most reconstruction networks are trained on post-ISP sRGB images, even though the image-signal-processing…
Jailbreak Attack Initializations as Extractors of Compliance Directions
Amit Levi, Rom Himelstein, Yaniv Nemcovsky +2
Safety-aligned LLMs respond to prompts with either compliance or refusal, each corresponding to distinct directions in the model's activation space. Recent works show that initiali…
Sparse patches adversarial attacks via extrapolating point-wise information
Yaniv Nemcovsky, Avi Mendelson, Chaim Baskin
Sparse and patch adversarial attacks were previously shown to be applicable in realistic settings and are considered a security risk to autonomous systems. Sparse adversarial pertu…
Hysteresis Activation Function for Efficient Inference
Moshe Kimhi, Idan Kashani, Avi Mendelson +1
The widely used ReLU is favored for its hardware efficiency, {as the implementation at inference is a one bit sign case,} yet suffers from issues such as the ``dying ReLU'' problem…