SWEAR: Section-Wise Efficient Audio Reasoning via Self-Rated Difficulty
On the map: LLM reasoning → Audio-language model reasoning
Abstract (arXiv)
Recent work on large audio-language models (LALMs) explores adaptive reasoning to balance accuracy and efficiency. However, existing approaches often rely on coarse-grained difficulty estimation, overlooking the diverse factors that determine reasoning effort. We address this limitation by deriving five-dimensional difficulty scores and compressed reasoning traces from complementary offline assessments. Based on this characterization, we propose SWEAR, a difficulty-aware reasoning framework that learns difficulty signals during supervised fine-tuning and uses reinforcement learning to adapt reasoning allocation across reasoning stages. Experiments on MMAU-test-mini and MMAR show that SWEAR improves accuracy while substantially reducing generated tokens, shortening reasoning for easy instances without sacrificing performance on difficult ones.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Audio-language model reasoning, one of 17 directions in LLM reasoning. Builds on 19 of its 22 references that are in the map. First posted on arXiv 9 Oct 2026.
Builds on
19 of 22 references.
Closest ancestors · 9
Same direction, newest first
- 2025Step-Audio-R1 Technical Report19 Nov
- 2025Incentivizing Consistent, Effective and Scalable Reasoning Capability in Audio LLMs via Reasoning Process Rewards23 Oct
- 2025Think Smart, Not Hard: Difficulty Adaptive Reasoning for Large Audio Language Models26 Sep
- 2025Omni-CLST: Error-aware Curriculum Learning with guided Selective chain-of-Thought for audio question answering14 Sep
- 2025Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models10 Jul
- 2025SARI: Structured Audio Reasoning via Curriculum-Guided Reinforcement Learning22 Apr
- 2025Qwen2.5-Omni Technical Report26 Mar
- 2025Audio-Reasoner: Improving Reasoning Capability in Large Audio Language Models4 Mar
- 2025Audio-CoT: Exploring Chain-of-Thought Reasoning in Large Audio Language Model13 Jan
Other references · 10
- 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
- 2025Qwen3-Omni Technical Reportcited by 497
- 2025DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Modelscited by 158
- 2025AdaptThink: Reasoning Models Can Learn When to Thinkcited by 147
- 2025Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RLcited by 62
- 2025Omni-R1: Do You Really Need Audio to Fine-Tune Your Audio LLM?cited by 46
- 2025Fast-Slow Thinking GRPO for Large Vision-Language Model Reasoningcited by 36
- 2025Prolonged Reasoning Is Not All You Need: Certainty-Based Adaptive Routing for Efficient LLM/MLLM Reasoningcited by 35
- 2025AdaCuRL: Adaptive Curriculum Reinforcement Learning with Invalid Sample Mitigation and Historical Revisitingcited by 20
- 2025Omni-AutoThink: Adaptive Multimodal Reasoning via Reinforcement Learningcited by 5
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models
- 2025Audio-Thinker: Guiding Audio Language Model When and How to Think via Reinforcement Learning
- 2026AuTAgent: A Reinforcement Learning Framework for Tool-Augmented Audio Reasoning
- 2026When Scaling Fails: Mitigating Audio Perception Decay of LALMs via Multi-Step Perception-Aware Reasoning
- 2026AudioRouter: Data Efficient Audio Understanding via RL based Dual Reasoning
- 2026Audio-DeepThinker: Progressive Reasoning-Aware Reinforcement Learning for High-Quality Chain-of-Thought Emergence in Audio Language Models
- 2026Listen Then Reason: Perception-Grounded Test-Time Reinforcement Learning for Large Audio-Language Models
- 2026CORD: Bridging the Audio-Text Reasoning Gap via Weighted On-policy Cross-modal Distillation