Auditing Future-Memory Use in Point-in-Time Language Agents
Point-in-time (PIT) agents can recall post-cutoff facts even when their external evidence respects a historical cutoff.
ProofPaper ↗
Key points
- We examine whether this access predicts evidence gathering in 48 historical company-comparison tasks.
- Independent closed-book revenue probes identify the better-remembered company for each model on tasks with unequal recall, fixing the predicted direction before agent runs.
- We then compare paired runs with real or anonymized company identities while keeping external evidence PIT-valid.
- The mechanism behind the natural inverse pattern remains unresolved.
Sources (1)
- [1]Auditing Future-Memory Use in Point-in-Time Language AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 13, 03:56 PM
Point-in-time (PIT) agents can recall post-cutoff facts even when their external evidence respects a historical cutoff.
We examine whether this access predicts evidence gathering in 48 historical company-comparison tasks.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 13, 2026black-forest-labs/flux-3-image-klein-comfy
- Oct 13, 2026MHVF: Training-Free GUI Grounding via Verification-Guided Multi-Hypothesis Search
- Oct 13, 2026Hybrid Reasoning Agents
- Oct 12, 2026Separating Prompt Harmfulness from Interaction Safety via Partial Generation
- Oct 12, 2026RelLedger: Relational Information Governance for LLM Agent Organizations
- Oct 12, 2026Hashrope: A Persistent Data Structure for Mechanisms of LLM Context Management