Pith. sign in

Researcher Evidence Record

Xiangzheng Zhang

This bounded record lists 32 Pith paper rows and 0 imported work rows attributed to this corpus identity. The enumerated, non-disputed paper rows include cs.CL, cs.AI, cs.CR work dated 2022 to 2026. The record describes sources and coverage; it makes no judgment about the person.

Compiled coverage vector

Measured lane counts only. Not a trust score or person verdict.

Enumerated paper scope: 6 fields (cs.CL, cs.AI, cs.CR, +3 more) · 2022-2026 sources: authors, author_identifiers · paper_authors · author_works · current_verdicts · cited_works

A sourced case file for attributed work. It is neither a profile score nor a verdict about this researcher.

Attributed works

A bounded ledger from the Pith paper and imported-work queries. Counts and source confidence stay with each work.

  1. 2026 Pith paper

    SkillSentry: Adaptive Honey Worlds for Dynamic Safety Testing of Agent Skills

    cs.CR provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    10
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  2. 2026 Pith paper

    SafeFlow: Semantic Information-Flow Control for Blocking Malicious Propagation in Multi-Agent Systems

    cs.MA provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    13
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  3. 2026 Pith paper

    Dynamic Defense Profiling Enables Cognitive Jailbreak of Text-to-Image Models

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  4. 2026 Pith paper

    ElephantAgent: Contextual State Continuity in Agentic Systems

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    2
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  5. 2026 Pith paper

    External Experience Serving in Production LLM Systems: A Deployment-Oriented Study of Quality-Cost Trade-offs

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    3
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  6. 2026 Pith paper

    RealClawBench: Live OpenClaw Benchmarks from Real Developer-Agent Sessions

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  7. 2026 Pith paper

    A Primer in Post-Training Reasoning Data: What We Know About How It Works

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    5
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 0 external cited by from pith
    • 1 pith inbound references from cited_work_pith_inbound_counts
  8. 2026 Pith paper

    Harness-Bench: Measuring Harness Effects across Models in Realistic Agent Workflows

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    11
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 2 pith inbound references from cited_work_pith_inbound_counts
  9. 2026 Pith paper

    MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    9
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  10. 2026 Pith paper

    DMN: A Compositional Framework for Jailbreaking Multimodal LLMs with Multi-Image Inputs

    cs.CR provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    6
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  11. 2026 Pith paper

    Demystifying and Detecting Agentic Workflow Injection Vulnerabilities in GitHub Actions

    cs.CR provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 0 external cited by from pith
    • 1 pith inbound references from cited_work_pith_inbound_counts
  12. 2026 Pith paper

    SafeHarbor: Defining Precise Decision Boundaries via Hierarchical Memory-Augmented Guardrail for LLM Agent Safety

    cs.CR provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 1 pith inbound references from cited_work_pith_inbound_counts
  13. 2026 Pith paper

    TrajShield: Trajectory-Level Safety Mediation for Defending Text-to-Video Models Against Jailbreak Attacks

    cs.CV provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    6
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  14. 2026 Pith paper

    When Good OCR Is Not Enough: Benchmarking OCR Robustness for Retrieval-Augmented Generation

    cs.CV provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  15. 2026 Pith paper

    Thinking with Reasoning Skills: Fewer Tokens, More Accuracy

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    4
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 0 external cited by from pith
    • 2 pith inbound references from cited_work_pith_inbound_counts
  16. 2026 Pith paper

    BEAR: Budgeted Evidence Allocation for Multi-Document Reasoning

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    6
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  17. 2026 Pith paper

    Robust Privacy: Inference-Stage Privacy through Certified Robustness

    cs.LG provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    2
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  18. 2025 Pith paper

    Disentangling Fact from Sentiment: A Dynamic Conflict-Consensus Framework for Multimodal Fake News Detection

    cs.LG provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    10
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  19. 2025 Pith paper

    TextlessRAG: End-to-End Visual Document RAG by Speech Without Text

    cs.CV provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    6
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 1 pith inbound references from cited_work_pith_inbound_counts
  20. 2025 Pith paper

    Mask-GCG: Are All Tokens in Adversarial Suffixes Necessary for Jailbreak Attacks?

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    9
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 2 pith inbound references from cited_work_pith_inbound_counts
  21. 2025 Pith paper

    Router Upcycling: Leveraging Mixture-of-Routers in Mixture-of-Experts Upcycling

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    9
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 1 pith inbound references from cited_work_pith_inbound_counts
  22. 2025 Pith paper

    Uncertainty Under the Curve: A Sequence-Level Entropy Area Metric for Reasoning LLM

    cs.AI provisional current review present

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    5
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  23. 2025 Pith paper

    Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following

    cs.CL provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    4
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  24. 2025 Pith paper

    PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking

    cs.CR provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    10
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 2 pith inbound references from cited_work_pith_inbound_counts
  25. 2025 Pith paper

    Evaluation is All You Need: Strategic Overclaiming of LLM Reasoning Capabilities Through Evaluation Design

    cs.AI provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    11
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 2 pith inbound references from cited_work_pith_inbound_counts
  26. 2025 Pith paper

    Light-R1: Curriculum SFT, DPO and RL for Long COT from Scratch and Beyond

    cs.CL provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    14
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 0 external cited by from arxiv_reference
    • 20 pith inbound references from cited_work_pith_inbound_counts
  27. 2025 Pith paper

    Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application

    cs.CR provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    7
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 3 pith inbound references from cited_work_pith_inbound_counts
  28. 2025 Pith paper

    TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation

    cs.CL provisional current review present

    Sources and evidence
    Authorship source
    backfill
    Printed name
    Xiangzheng Zhang
    Author position
    21
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: a current Pith review exists.
    Citation counts
    • 2 pith inbound references from cited_work_pith_inbound_counts
  29. 2025 Pith paper

    Chain-of-Thought Matters: Improving Long-Context Language Models with Reasoning Path Supervision

    cs.CL provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    9
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 1 pith inbound references from cited_work_pith_inbound_counts
  30. 2025 Pith paper

    Reasoning-Augmented Conversation for Multi-Turn Jailbreak Attacks on Large Language Models

    cs.CL provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    8
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 6 pith inbound references from cited_work_pith_inbound_counts
  31. 2024 Pith paper

    Expand VSR Benchmark for VLLM to Expertize in Spatial Rules

    cs.CV provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    5
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    No source count is attached to this work row.
  32. 2022 Pith paper

    CCMB: A Large-scale Chinese Cross-modal Benchmark

    cs.CV provisional measured, no current review

    Sources and evidence
    Authorship source
    arxiv_oai
    Printed name
    Xiangzheng Zhang
    Author position
    10
    Identity state
    provisional
    Source confidence
    0.7
    Review coverage
    Measured: no current Pith review exists.
    Citation counts
    • 1 pith inbound references from cited_work_pith_inbound_counts

Evidence apparatus

The machinery behind this record. Every lane states whether Pith measured it, did not query it, could not reach it, or withheld it.

LaneStateObservedBoundary and source
identity Measured 2 Canonical identity row plus public typed identifiers.
source=authors, author_identifiers
papers Measured 32 of 32 bounded rows Rows attributed to this author UUID in the Pith corpus.
source=paper_authors
works Measured zero 0 of 0 bounded rows Imported works not duplicated by the paper ledger.
source=author_works
reviews Measured 23 of 32 bounded rows Coverage count only. No review outcome is projected onto the person.
source=current_verdicts
citations Measured 16 of 32 bounded rows Counts remain itemized by work and source.
source=cited_works
coauthors Measured 50 of 32 bounded rows Shared-work edges from admitted paper rows.
source=paper_authors
account Unavailable No public count of 1 bounded rows Account metadata is separate from corpus evidence.
source=users.author_id
Public identity sources
  • name variant
    Xiangzheng Zhang
    backfill
    confidence 0.6
Enumerated research scope

Fields and dates come only from enumerated, non-disputed Pith paper rows. They do not claim career completeness.

  • cs.CL11 rows
  • cs.AI7 rows
  • cs.CR6 rows
  • cs.CV5 rows
  • cs.LG2 rows
  • cs.MA1 rows
  • 20221 rows
  • 20241 rows
  • 202513 rows
  • 202617 rows
Shared-work index
Record scope

The work queries are bounded. Missing rows may mean measured zero, an unavailable source, a query that did not run, or private data that Pith withheld. The lane table keeps those cases separate.

Paper findings remain attached to papers. They do not become findings about this researcher.