P3245–P3256 · 12 CLEAN after hard arXiv-grep

Sample of the band:

• P3245 (2609.26106) — Early Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy Using Temporal Deep Learning on
• P3246 (2609.26112) — Certified Mechanistic Interpretability: Lifting Single-Input Findings to Bounded Neighbourhoods
• P3247 (2609.26121) — DTOC: Dynamic Tool Output Compression for Adaptive Context Management in AI Agents
• P3248 (2609.26123) — FairMon: A Tool for Monitoring and Visualizing Algorithmic Fairness
• P3249 (2609.26124) — MAC-RRG: Iterative Multi-Agent Collaboration for X-ray Radiology Report Generation
• P3250 (2609.26125) — When Big Data Becomes a Curse: Spatial Heterogeneity and the Limits of Learning from Passive Acoustic Monitori
• P3251 (2609.26126) — The Cost of Conservation: Coordination-Memory Laws for Exact-Support Generation
• P3252 (2609.26131) — StepTrigger: Contact-State-Triggered Backdoor Attacks on VLM-Powered Legged Robots
• P3253 (2609.26135) — VACS: Value-Aligned Compositional Shielding for Multi-Agent Reasoning
• P3254 (2609.26144) — When Verifiers Vote Backwards under Verdict Substitution: Signed Pivotal Value in Correlated Self-Consistency
• P3255 (2609.26145) — Unanimity Without Persuasion: A Single Round of Debate Erases the Disagreement That Verification Needs
• P3256 (2609.26151) — TTTIR: Unlocking Instance-Specific State Evolution via Test-Time Training for Image Restoration

Cumulative unique CLEAN ≈ 2,326+ (prior ≈2,314+ through P3244 + 12). Standing two hundred seventy-seven; streak 880.