TRAM: Enhancing Multimodal Reasoning with Trajectory-Derived Auxiliary Memory

arXiv:2608.01922 · cs.CL · Submitted 2026-08-03 · Read on arXiv

Kang Liu, Zijing Wang, Yongkang Liu, Mengjie Zhao, Xiaocui Yang, Shi Feng, Yifei Zhang, Daling Wang

cs.CL

Submitted: 2026-08-03

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 17× in this paper · explained in 2 episodes
• Trace — 14× in this paper · explained in 2 episodes
• Decoder — 13× in this paper · explained in 2 episodes
• Accuracy — 12× in this paper · explained in 1 episode
• Multimodal Reasoning — 12× in this paper · explained in 5 episodes
• Hallucination — 8× in this paper · explained in 4 episodes
• Grounding — 6× in this paper · explained in 3 episodes
• Ablation — 5× in this paper · explained in 1 episode
• Reinforcement Learning — 5× in this paper · explained in 4 episodes
• Computer Vision and Pattern Recognition — 3× in this paper
• Papers — 3× in this paper · explained in 3 episodes
• Steering — 3× in this paper · explained in 2 episodes
• Transformer — 3× in this paper · explained in 2 episodes
• Visual Grounding — 3× in this paper · explained in 2 episodes
• Calibration — 2× in this paper · explained in 9 episodes
• Faithfulness — 2× in this paper · explained in 2 episodes
• Transformers — 2× in this paper · explained in 2 episodes
• Vision-language models — 2× in this paper · explained in 2 episodes
• Ablation Study — 1× in this paper · explained in 3 episodes
• Attention Mechanism — 1× in this paper · explained in 3 episodes
• Clarity — 1× in this paper · explained in 3 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Hallucination Mitigation — 1× in this paper · explained in 2 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• RLVR — 1× in this paper · explained in 2 episodes
• Reasoning Traces — 1× in this paper · explained in 4 episodes
• Sparsity — 1× in this paper · explained in 5 episodes
• Test-Time Scaling — 1× in this paper · explained in 4 episodes
• Training-Free Framework — 1× in this paper · explained in 2 episodes
• Training-Free Intervention — 1× in this paper · explained in 2 episodes
• Transformer Architecture — 1× in this paper · explained in 4 episodes
• Verifiable Rewards — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• Qwen3-VL Technical Report
• Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs
• TransformerFAM: Feedback attention is working memory
• Memory-Augmented Transformers: A Systematic Review from Neuroscience Principles to Enhanced Model Architectures
• Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs
• OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning
• InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering