Searcharxiv⌕ Search

arXiv subjects

Junyuan Tan

Publications and source records attributed to Junyuan Tan.

4 recordsLinked to original sources

Where the Evidence Lives: Measuring Privacy and Diagnosability of Exported AI-Agent Traces

Execution traces help diagnose AI-agent failures, but sharing them can expose customer identities. For example, detecting that an agent handled the wrong customer's records requires knowing whether the records belong to the requested customer, but not necessarily knowing who that customer is. Our executable benchmark measures what a trace export reveals and whether a proposed repair restores the correct output in synthetic multi-tenant workflows. Conventional redaction can remove evidence needed for repair while leaving encoded identifiers recoverable. Two exports preserve ownership comparisons by reporting the match as a comparison fact or by using pseudonyms consistent within each execution. Both preserve the original diagnostic program's repairs. The tested identifier matchers identify no requested customer from either export. An independently written program produces correct outputs on 39 of 42 faulty cases with comparison facts and on all 42 with execution-scoped pseudonyms. Row counts still expose identities. With prior knowledge of customers' row counts, an attacker uniquely identifies the requested customer in 97 of 240 executions under each of the 15 export policies. Removing selected counters from the comparison-fact export stops this attack and preserves the original program's repairs. The benchmark lets engineers test what their diagnostic program can repair and what an attacker can still infer from the same export.

cs.CR↗

Who May Release, Who Must Repair: Separating Authorization from Report Accountability in Personalized Services

Personalized services often rely on different parties to compute results and decide whether data permissions and service policies allow delivery. The computing party's report may omit permitted inputs, leaving the delivery party to decide whether to wait for correction. Requiring correction first can delay service despite valid input permissions. Delivering first instead requires preserving responsibility for the incomplete report. We design an interface that checks input permissions separately from reporting requirements and records who must correct omissions and who may verify corrections. We evaluate three implementations on scripted histories and generated cases against a design treating omissions as authorization failures. When policy permits delivery before correction, our interface releases results that the alternative withholds while preserving repair assignments. These assignments remain in force through policy changes, so changing delivery rules does not erase earlier obligations. The interface makes the basis for release and outstanding repair responsibilities traceable across cooperating parties.

cs.CY↗

RoboMemory: A Brain-inspired Multi-memory Agentic Framework for Interactive Environmental Learning in Physical Embodied Systems

Long-horizon embodied tasks require agents to learn, reason, and generalize robustly across complex real-world environments. However, existing approaches often struggle with partial observability, fragmented spatial reasoning, and inefficient integration of heterogeneous memories, limiting their capacity for long-horizon adaptation. To address this, we introduce RoboMemory, a brain-inspired framework that unifies Temporal, Spatial, Semantic, and Episodic memory for efficient long-horizon planning and interactive learning. Its core innovations are a comprehensive memory system designed for embodied tasks, and a dynamic spatial knowledge graph for scalable, consistent memory updates. Extensive experiments on EmbodiedBench show that RoboMemory improves the average success rate of Qwen2.5-VL-72B-Ins by 26.5 percentage points and achieves performance comparable to strong closed-source VLM agents. Real-world trials further suggest RoboMemory's potential for interactive environmental learning by reusing accumulated experience across repeated task attempts. These results demonstrate RoboMemory as a scalable memory-augmented framework for embodied agents, improving long-horizon planning and cumulative learning in complex interactive environments.

cs.RO↗

Verification-Notebook Learning for Source-Aware Multimodal Misinformation Detection

Multimodal misinformation verification is challenging because misleading signals may come from different parts of a post and require different forms of evidence. LVLMs are well suited to this task, but their verification performance often depends on the inference procedure applied to each instance. Existing methods improve this procedure through stronger prompting, retrieval, or deliberation, but rarely retain the verification patterns learned from previous examples. We propose Verification-Notebook Learning (VNL), a non-parametric framework that learns an external verification procedure for a frozen LVLM before inference. VNL builds a compact notebook of decision principles, evidence cues, and recurring pitfalls from prior verification experience. The notebook remains fixed during inference and guides the verification of new examples. Rather than updating model parameters or storing demonstrations, VNL records learned knowledge in an artifact that can be inspected directly. Experiments show that VNL consistently outperforms a range of competitive baselines. Further analyses show that the Verification Notebook improves fine-grained source attribution while remaining compact and interpretable, providing an effective way to accumulate verification knowledge without model training.

cs.AI↗