Roy Ganz

Switching between models incurs a "handoff tax" that can significantly degrade quality while increasing costs, challenging the assumption that stronger models always yield better outcomes.

Richer trace representations can dramatically enhance failure attribution in multi-agent systems, achieving new benchmarks in accuracy.

Static benchmarks can be fooled by fluent text and aligned citations, but DREAM leverages agentic evaluation to expose the critical capability mismatch in assessing temporal validity and factual correctness of research agents.

Ditch slow, token-by-token OCR: DODO unlocks 3x faster inference by reframing OCR as a parallelizable diffusion process.