Shahar Katz

7 papers · 2023–2026 · 5 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Silver 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (13) 🐣 Hot Topic Early Bird 🏆 Keyword Champion (2)

Venues

ACL (2) EMNLP (2) AAAI (1) EACL (1) NAACL (1)

Top co-authors

Lior Wolf (6) Yonatan Belinkov (2) Ariel Shaulov (1) Yaniv Romano (1) Bar Alon (1) Itamar Zimerman (1) Mor Geva (1) Liran Ringel (1) Mahmood Sharif (1) Ido Andrew Atad (1)

Research topics

Artificial Intelligence (6) Deep Learning (6) Core AI (6) Optimization & Theory (4) Architectures (3) Natural Language Processing (2) Generation (2) Machine Learning (1) Core Methods (1) Learning Types (1) Resources & Methods (1) Techniques (1)

Keywords

attention mechanism (3) language model (2) generative pre-trained transformer (2) jailbreak attack (2) gradient descent (1) support vector machine (1) hidden state (1) model fine-tuning (1) attention head (1) hidden state analysis (1) attention masking (1) linear representation (1) information flow (1) gradient analysis (1) random forest classifier (1) causal masking (1) prefill phase (1) transformer interpretability (1) weight modification (1) vocabulary space (1)

Papers

AlignTree: Efficient Defense Against LLM Jailbreak Attacks AAAI 2026 Safeguarding Language Models via Self-Destruct Trapdoor EACL 2026 TensorLens: End-to-End Transformer Analysis via High-Order Attention Tensors ACL 2026 Reversed Attention: On The Gradient Descent Of Attention Layers In GPT NAACL 2025 Segment-Based Attention Masking for GPTs ACL 2025 Backward Lens: Projecting Language Model Gradients into the Vocabulary Space EMNLP 2024 VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers EMNLP 2023