Lattice AI Research

Research focus

Architecture Design (Transformers, SSMs, MoE) (1)Inference & Quantization (1)Training Efficiency & Optimization (1)Code Generation & Program Synthesis (1)

Frequent co-authors

Alon Helvits (1)Noam Tarshish (1)Nofar Selouk (1)Daniel Hodisan (1)

Papers (2)

May 27, 2026

Alon Helvits +1May 27, 2026

Score Based Error Correcting Code Decoder

Ditch complex SNR estimation: this new score-based decoder learns to correct errors across noise levels, outperforming existing methods by 0.17dB on average.

Alon Helvits, Eliya Nachmani

Architecture Design (Transformers, SSMs, MoE)Inference & Quantization Training Efficiency & Optimization

Apr 28, 2026

Ben-Gurion University of the NegevApr 28, 2026

SAFEdit: Does Multi-Agent Decomposition Resolve the Reliability Challenges of Instructed Code Editing?

Multi-agent code editing with structured failure feedback boosts task success by 17%, suggesting a promising path to more reliable LLM-driven code manipulation.

Noam Tarshish, Nofar Selouk, Daniel Hodisan +4

Code Generation & Program Synthesis Eval Frameworks & Benchmarks Tool Use & Agents

Search

Eliya Nachmani

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)