Apple paper: structured selection-based reasoning cuts search agent latency by 90%
_reachsumit · x · 2026-10-02
- Apple researchers published SSR (Selection-based Structured Reasoning), which replaces free-form reasoning in multimodal search agents with selection from pre-specified, reusable natural-language reasoning candidates based on token likelihoods.
- Pre-defined reasoning traces enable parallel scoring via teacher-forced prefilling with a shared context KV cache, no auxiliary task head needed.
- Evaluated on seven multimodal search benchmarks with 2B and 4B models across RL objectives and SFT, SSR matches the success rate of leading same-scale search agents while cutting per-turn reasoning latency by over 90% and total per-question inference latency substantially.
More from Research
- Stanford's Le Cong to talk AI co-scientists for RNA design at 14th mRNA Health Conference — MengdiWang10 · 2026-11-19
- New paper shows DiT latents can be linearly probed to predict pixels, skipping the VAE decoder — kwangmoo_yi · 2026-11-19
- SAIL 2027 opens abstract submissions for AI in clinical care, deadline Jan 15 — zakkohane · 2026-11-19
- The academia LLM paradox: everyone writes with AI, nobody wants to read or be judged by it — wandedob · 2026-11-19
- Baseten hosts All-Gather, its first research conference on where AI's next gains will come from — baseten · 2026-11-19
- Interpretability solved—then what? Najoung Kim proposes filtering for recurrent constructs — najoungkim · 2026-11-19