AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization

• URL: http://arxiv.org/abs/2606.14694v2
• Date: Mon, 15 Jun 2026 16:28:15 GMT
• Title: AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization
• Abstract summary: AdaSR is an adaptive streaming reasoning framework that enables models to reason during input streaming and perform final deliberation once the stream is complete.<n>HRPO integrates format, accuracy, and adaptive thinking rewards to enforce valid reasoning protocols, preserve final task performance, and encourage latency-aware computation allocation.<n> Experiments show that AdaSR achieves a better balance among reasoning accuracy, computational efficiency, and streaming latency compared with supervised fine-tuning baseline.
• Score: 12.363024790235158
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large reasoning models typically follow a read-then-think paradigm: they observe the complete input, reason over a static context, and then produce the answer. Yet many real-world scenarios are inherently dynamic, such as audio and video stream, where information arrives as a continuous stream and models must reason, update, and respond under partial observations. Recent streaming reasoning methods allow models to think while reading, but they largely rely on supervised imitation of pre-constructed trajectories, which limits their flexibility. In this paper, we propose AdaSR, an adaptive streaming reasoning framework that enables models to reason during input streaming and perform final deliberation once the stream is complete, learning when to think, and how much computation to allocate across different stages. To optimize this hierarchical reasoning process, we introduce Hierarchical Relative Policy Optimization (HRPO), which decomposes policy optimization into streaming reasoning and deep reasoning phases, providing more fine-grained advantage assignment instead of uniformly distributing a single sequence-level advantage over all tokens. HRPO integrates format, accuracy, and adaptive thinking rewards to enforce valid reasoning protocols, preserve final task performance, and encourage latency-aware computation allocation. Experiments show that AdaSR achieves a better balance among reasoning accuracy, computational efficiency, and streaming latency compared with supervised fine-tuning baseline. We release our code at https://github.com/EIT-NLP/StreamingLLM/tree/main/AdaSR.

Related papers

• CARE: Competence-Aware Reward Shaping for Adaptive Reasoning Length in Video-MLLMs [50.189987475377656]
  We propose CARE, a competence-aware reward shaping framework for adaptive reasoning length optimization in multimodal reasoning.<n> CARE maintains a smoothed competence estimate via an exponential moving average of pass rates, and uses it to route training into progressive stages.<n>Experiments on multiple video reasoning and general video understanding benchmarks demonstrate that CARE consistently improves reasoning accuracy, stabilizes reinforcement learning, and significantly enhances token efficiency.
  arXiv  Detail & Related papers  (2026-06-18T08:28:26Z)
• Reasoning Matters: Mitigate Hallucination in Multimodal Large Reasoning Models via Reasoning-Conditioned Preference Optimization [59.20570719781289]
  Multimodal Large Reasoning Models introduce the reasoning paradigm, demonstrating strong capabilities on complex vision-language tasks.<n>Existing training-based methods typically mitigate hallucinations through response-level direct preference optimization (DPO), where the Chain-of-Thought (CoT) and the final answer are treated as a monolithic output and optimized jointly.<n>We derive Reasoning-Conditioned Direct Preference Optimization (RC-DPO), which models the CoT as a condition for answer generation and contrasts the preference for the same preferred answer under different CoT conditions, promoting answer-supportive reasoning chain alignment.
  arXiv  Detail & Related papers  (2026-05-27T03:27:23Z)
• StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video [53.82672457255517]
  Proactive streaming video understanding requires models to continuously process video streams and decide when to respond, rather than merely what to respond.<n>Existing benchmarks largely follow a "see-then-answer" paradigm, where responses are triggered only after explicit evidence appears.<n>We introduce StreamPro, a new benchmark that evaluates streaming models from three complementary perspectives: Perception Understanding, Temporal Reasoning, and Proactive Agency.
  arXiv  Detail & Related papers  (2026-05-11T05:01:15Z)
• Thinking in Streaming Video [30.61790766076081]
  ThinkStream is a framework for streaming video reasoning based on a Watch--Think--Speak paradigm.<n> Reasoning-Compressed Streaming Memory (RCSM) treats intermediate reasoning traces as compact semantic memory.<n>Experiments on multiple streaming video benchmarks show that ThinkStream significantly outperforms existing online video models.
  arXiv  Detail & Related papers  (2026-03-13T12:33:36Z)
• Think-as-You-See: Streaming Chain-of-Thought Reasoning for Large Vision-Language Models [14.21980212001207]
  Motivated by the streaming nature of video data, we investigate two streaming reasoning paradigms for LVLMs.<n>To better match streaming inputs, we propose textbfThink-as-You-See (TaYS), a unified framework enabling true concurrent reasoning.
  arXiv  Detail & Related papers  (2026-03-03T11:24:55Z)
• The Path of Least Resistance: Guiding LLM Reasoning Trajectories with Prefix Consensus [1.7798494187119376]
  We introduce PoLR, the first inference-time method to leverage prefix consistency for compute-efficient reasoning.<n>PoLR clusters short prefixes of reasoning traces, identifies the dominant cluster, and expands all paths in that cluster.<n> Empirically, PoLR consistently matches or exceeds SC across GSM8K, MATH500, AIME24/25, and GPQA-DIAMOND.
  arXiv  Detail & Related papers  (2026-01-29T10:14:24Z)
• StreamingCoT: A Dataset for Temporal Dynamics and Multimodal Chain-of-Thought Reasoning in Streaming VideoQA [60.86024022291499]
  We introduce StreamingCoT, the first dataset explicitly designed for temporally evolving reasoning in streaming video streams.<n>Our framework generates per-second dense descriptions and constructs temporally-dependent semantic segments through similarity fusion.<n>This dataset establishes a foundation for advancing research in streaming video understanding, complex temporal reasoning, and multimodal inference.
  arXiv  Detail & Related papers  (2025-10-29T09:47:38Z)
• StreamingThinker: Large Language Models Can Think While Reading [14.54868327561777]
  Large language models (LLMs) have demonstrated remarkable capabilities in chain of thought (CoT) reasoning.<n>Inspired by human cognition of thinking while reading, we first design a textittextbfstreaming thinking paradigm for LLMs.<n>We instantiate this paradigm with textitStreamingThinker, a framework that enables LLMs to think while reading.
  arXiv  Detail & Related papers  (2025-10-20T07:27:37Z)
• HiPO: Hybrid Policy Optimization for Dynamic Reasoning in LLMs [54.16300997612526]
  Large Language Models (LLMs) increasingly rely on Chain-of-Thought (CoT) reasoning to improve accuracy on complex tasks.<n>This paper introduces the Hybrid Policy Optimization (i.e., HiPO), a framework for adaptive reasoning control.<n> Experiments across mathematics and coding benchmarks demonstrate that HiPO can substantially reduce token length while maintaining or improving accuracy.
  arXiv  Detail & Related papers  (2025-09-28T16:46:12Z)
• SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought [37.53215651690168]
  Chain of Thought (CoT) prompting improves the reasoning performance of large language models (LLMs) by encouraging step by step thinking.<n>While promising, CoT-based approaches often require costly pretraining and lack a principled framework for how reasoning should evolve.<n>We propose SCOUT, a lightweight fine tuning framework that enables Flow CoT style reasoning without the need for pretraining.
  arXiv  Detail & Related papers  (2025-05-30T03:43:24Z)
• PixelThink: Towards Efficient Chain-of-Pixel Reasoning [70.32510083790069]
  PixelThink is a simple yet effective scheme that integrates externally estimated task difficulty and internally measured model uncertainty.<n>It learns to compress reasoning length in accordance with scene complexity and predictive confidence.<n> Experimental results demonstrate that the proposed approach improves both reasoning efficiency and overall segmentation performance.
  arXiv  Detail & Related papers  (2025-05-29T17:55:49Z)
• Non-Cooperative Game Theory Based Rate Adaptation for Dynamic Video Streaming over HTTP [89.30855958779425]
  Dynamic Adaptive Streaming over HTTP (DASH) has demonstrated to be an emerging and promising multimedia streaming technique. We propose a novel algorithm to optimally allocate the limited export bandwidth of the server to multi-users to maximize their Quality of Experience (QoE) with fairness guaranteed.
  arXiv  Detail & Related papers  (2019-12-27T01:19:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.