Moshe Berchansky

15 papers A* 3Journal 8Unranked 4

Year	Rank	Type	Title / Venue / Authors
2025	A*	`conf`	Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies. ICML Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Gaurav Jain, Oren Pereg, Moshe Wasserblat, David Harel
2025	J	`jnl`	Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies. CoRR Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Oren Pereg, Gaurav Jain, Roy Schwartz, Moshe Wasserblat, David Harel
2025	A*	`conf`	Distributed Speculative Inference (DSI): Speculation Parallelism for Provably Faster Lossless Language Model Inference. ICLR Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Oren Pereg, Moshe Wasserblat, Tomer Galanti, Michal Gordon-Kiwkowitz, David Harel
2025	J	`jnl`	SQuARE: Sequential Question Answering Reasoning Engine for Enhanced Chain-of-Thought in Large Language Models. CoRR Daniel Fleischer, Moshe Berchansky, Gad Markovits, Moshe Wasserblat
2024	J	`jnl`	Accelerating Speculative Decoding using Dynamic Speculation Length. CoRR Jonathan Mamou, Oren Pereg, Daniel Korat, Moshe Berchansky, Nadav Timor, Moshe Wasserblat, Roy Schwartz
2024	—	`conf`	CoTAR: Chain-of-Thought Attribution Reasoning with Multi-level Granularity. EMNLP (Findings) Moshe Berchansky, Daniel Fleischer, Moshe Wasserblat, Peter Izsak
2024	J	`jnl`	CoTAR: Chain-of-Thought Attribution Reasoning with Multi-level Granularity. CoRR Moshe Berchansky, Daniel Fleischer, Moshe Wasserblat, Peter Izsak
2024	—	`conf`	Distributed Speculative Inference of Large Language Models is Provably Faster. ENLSP Nadav Timor, Jonathan Mamou, Oren Pereg, Moshe Berchansky, Daniel Korat, Moshe Wasserblat, Tomer Galanti, Michal Gordon, David Harel
2024	J	`jnl`	Distributed Speculative Inference of Large Language Models. CoRR Nadav Timor, Jonathan Mamou, Daniel Korat, Moshe Berchansky, Oren Pereg, Moshe Wasserblat, Tomer Galanti, Michal Gordon, David Harel
2024	—	`conf`	Dynamic Speculation Lookahead Accelerates Speculative Decoding of Large Language Models. ENLSP Jonathan Mamou, Oren Pereg, Daniel Korat, Moshe Berchansky, Nadav Timor, Moshe Wasserblat, Roy Schwartz
2024	J	`jnl`	RAG Foundry: A Framework for Enhancing LLMs for Retrieval Augmented Generation. CoRR Daniel Fleischer, Moshe Berchansky, Moshe Wasserblat, Peter Izsak
2023	A*	`conf`	Optimizing Retrieval-augmented Reader Models via Token Elimination. EMNLP Moshe Berchansky, Peter Izsak, Avi Caciularu, Ido Dagan, Moshe Wasserblat
2023	J	`jnl`	Optimizing Retrieval-augmented Reader Models via Token Elimination. CoRR Moshe Berchansky, Peter Izsak, Avi Caciularu, Ido Dagan, Moshe Wasserblat
2021	—	`conf`	How to Train BERT with an Academic Budget. EMNLP (1) Peter Izsak, Moshe Berchansky, Omer Levy
2021	J	`jnl`	How to Train BERT with an Academic Budget. CoRR Peter Izsak, Moshe Berchansky, Omer Levy