SxS Interleaved Reasoning learns when to disclose partial reasoning during generation and improves accuracy versus content-latency trade-offs on math and science benchmarks.
However, independent LLM calls may produce noisy, non-monotonic results (e.g., but )
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
SxS Interleaved Reasoning learns when to disclose partial reasoning during generation and improves accuracy versus content-latency trade-offs on math and science benchmarks.