{"slug": "dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue", "title": "DMV-Bench: Diagnosing Long-Horizon Multimodal Agents' Visual Memory with Incidental Cue Injection", "summary": "Researchers introduced DMV-Bench, the first interactive benchmark for multimodal-agent visual memory, testing agents' ability to recall visual cues from product images in a simulated e-commerce environment. Their proposed DualMem architecture, which maintains parallel visual and verbal codes, outperformed existing systems across multiple chain lengths on Gemini 2.5 Flash and Qwen2.5-VL-7B. The benchmark addresses a gap in evaluating visual memory in interactive settings, where text-based memory has been the primary focus.", "body_md": "arXiv:2606.27499v1 Announce Type: new\nAbstract: Research on agent memory has matured rapidly, but almost entirely on the text side: few existing benchmarks ask, in an interactive environment, when an agent genuinely needs to remember what it saw rather than what it could write down. We introduce DMV-Bench (Code: https://github.com/yyyujintang/DMV-Bench), the first interactive benchmark for multimodal-agent visual memory. DMV-Bench is built on a controlled home-furnishing e-commerce catalogue of 1,000 product variants in which a text-leakage contract keeps the discriminative signal of each task in the pixels alone. Across a chain of autonomous shopping sessions, every visited product image carries a unique, pre-rendered incidental cue, and the agent is later asked to recall a particular cued product and navigate to its URL. Inspired by dual-coding theory, we propose DualMem, a memory architecture that maintains a visual and a verbal code in parallel. On DMV-Bench, DualMem outperforms a caption baseline and three recent multimodal agent-memory systems at every chain length J in {5, 10, 15, 50} on both Gemini 2.5 Flash and Qwen2.5-VL-7B, with the lead surviving controls for memory-bank size and encoding-position bias, and an asymmetric dual-coding regime in which vision carries the cue end-to-end while the verbal channel plays a smaller query-grounding role.", "url": "https://wpnews.pro/news/dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue", "canonical_source": "https://arxiv.org/abs/2606.27499", "published_at": "2026-06-29 04:00:00+00:00", "updated_at": "2026-06-29 04:02:39.426434+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-agents", "ai-research"], "entities": ["DMV-Bench", "DualMem", "Gemini 2.5 Flash", "Qwen2.5-VL-7B", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue", "markdown": "https://wpnews.pro/news/dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue.md", "text": "https://wpnews.pro/news/dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue.txt", "jsonld": "https://wpnews.pro/news/dmv-bench-diagnosing-long-horizon-multimodal-agents-visual-memory-with-cue.jsonld"}}