pith. sign in

Xiangzheng Zhang

Identifiers

  • name variant Xiangzheng Zhang 0.60 · backfill

Papers (10)

  1. MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection cs.AI · 2026 · author #9
  2. DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs cs.CR · 2026 · author #6
  3. Demystifying and Detecting Agentic Workflow Injection Vulnerabilities in GitHub Actions cs.CR · 2026 · author #7
  4. SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety cs.CR · 2026 · author #7
  5. TrajShield: Trajectory-Level Safety Mediation for Defending Text-to-Video Models Against Jailbreak Attacks cs.CV · 2026 · author #6
  6. When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation cs.CV · 2026 · author #7
  7. Thinking with Reasoning Skills: Fewer Tokens, More Accuracy cs.AI · 2026 · author #4
  8. Disentangling Fact from Sentiment: A Dynamic Conflict-Consensus Framework for Multimodal Fake News Detection cs.LG · 2025 · author #10
  9. PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking cs.CR · 2025 · author #10
  10. TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation cs.CL · 2025 · author #21

Mentions

  • 2605.23723 #9 · arxiv_oai · confidence 0.70 Xiangzheng Zhang
  • 2605.05704 #7 · arxiv_oai · confidence 0.70 Xiangzheng Zhang
  • 2605.18915 #6 · arxiv_oai · confidence 0.70 Xiangzheng Zhang

Frequent Coauthors