Lattice AI Research

Research focus

Red-Teaming & Adversarial Robustness (1)Training Efficiency & Optimization (1)Code Generation & Program Synthesis (1)Eval Frameworks & Benchmarks (1)

Frequent co-authors

Raya Elsaleh (1)Liam Davis (1)Guy Katz (1)Rocky Klopfenstein (1)

Papers (2)

Mar 12, 2026

Raya Elsaleh +34d ago

Incremental Neural Network Verification via Learned Conflicts

Neural network verification gets a speed boost: reusing learned conflicts across related queries cuts down redundant exploration and speeds up verification by up to 1.9x.

Raya Elsaleh, Liam Davis, Haoze Wu +1

Red-Teaming & Adversarial Robustness Training Efficiency & Optimization

Mar 4, 2026

Rocky Klopfenstein +51w ago

SpotIt+: Verification-based Text-to-SQL Evaluation with Database Constraints

Text-to-SQL evaluation gets a reality check: SpotIt+ uses LLM-validated database constraints to find discrepancies missed by standard testing.

Rocky Klopfenstein, Yang He, Andrew Tremante +3

Code Generation & Program Synthesis Eval Frameworks & Benchmarks Natural Language Processing

Search

Haoze Wu

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)