When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
Summary
This paper introduces Side-by-Side Interleaved Reasoning, a method for controlling disclosure timing in autoregressive models to improve accuracy and efficiency. It demonstrates improved performance on benchmarks using Qwen3 models by interleaving private reasoning with partial disclosures.
View Cached Full Text
Cached at: 05/08/26, 08:00 AM
Paper page - When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
Source: https://huggingface.co/papers/2605.03314
Abstract
Side-by-Side Interleaved Reasoning enables controlled disclosure timing in autoregressive models, improving accuracy and efficiency through interleaved private reasoning and delayed content release.
In single-streamautoregressive interfaces, the same tokens both update the model state and constitute an irreversible public commitment. This coupling creates asilence tax: additional deliberation postpones the first task-relevant content, while naive early streaming risks premature commitments that bias subsequent generations. We introduce Side-by-Side (SxS)Interleaved Reasoning, which makes disclosure timing a controllable decision within standard autoregressive generation. SxS interleaves partial disclosures with continuedprivate reasoningin the same context, but releases content only when it is supported by the reasoning so far. To learn such pacing without incentivizing filler, we construct entailment-aligned interleaved trajectories by matching answer prefixes to supporting reasoning prefixes, then train with SFT to acquire the dual-action semantics and RL to recover reasoning performance under the new format. Across two Qwen3 architectures/scales (MoEQwen3-30B-A3B, dense Qwen3-4B) and both in-domain (AIME25) and out-of-domain (GPQA-Diamond) benchmarks, SxS improves accuracy--content-latency Pareto trade-offs under token-level proxies such as inter-update waiting.
View arXiv pageView PDFAdd to collection
Get this paper in your agent:
hf papers read 2605\.03314
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2605.03314 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2605.03314 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2605.03314 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Selective Disclosure of Hidden Directives in Reasoning Models: Behavioral Asymmetry and Steering
This paper introduces the Instruction-Compliance Gap and finds behavioral asymmetry in reasoning models' disclosure of hidden directives, using steering vectors to manipulate this behavior.
Hidden Thoughts Are Not Secret: Reasoning Trace Exposure in LLMs
This paper introduces Reasoning Exposure Prompting (REP), a method that uses shadow-model demonstrations in code-like formats to elicit hidden reasoning traces from LLMs, showing that interface-level trace hiding is insufficient to prevent extraction of useful reasoning signals.
Second Thought: Reasoning in Parallel as LLM Agents Act and Observe
Second Thought is a training-free framework that runs auxiliary reasoning branches in parallel during LLM agent action-observation waits to reduce sequential decoding and turn counts without harming accuracy.
When to Think Deeply: Inhibitory Deliberation for LLM Reasoning
IDPR is a framework for response-conditioned inhibitory deliberation that first generates a fast intuitive answer, then uses an inhibition controller to decide whether to invoke slow reasoning, achieving efficiency gains while maintaining accuracy.
Adaptive Triggering for Bias Correction in LLM Reasoning
This paper introduces an adaptive triggering framework for bias correction in LLM chain-of-thought reasoning, using online change-point detection to optimize intervention timing with white-box and black-box signals, improving accuracy and reducing interventions.