ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
arXiv preprint arXiv:2404.16019 (2024)
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
other 3
background 1
citation-polarity summary
representative citing papers
PERMA is a new benchmark using temporally ordered events, text variability, and linguistic alignment to evaluate LLM memory agents on persona consistency beyond simple retrieval.
Crowdsourced metaphors show rising anthropomorphism and warmth toward AI that predict trust and adoption, with notable demographic differences.
RAC is a closed-form bias correction for delayed rewards in RLHF that is unbiased under full mass reinjection of the delay kernel and reduces to V-trace with no delay.
PEBS applies Morris-James-Stein empirical-Bayes shrinkage to per-rater affine calibrators in RLHF, cutting within-user held-out RMSE by 8.58% on PRISM and 9.66% on PluriHarms versus pooled baselines.
DISCA converts within-country disagreement among World Values Survey personas into a bounded logit correction that reduces cultural misalignment by 10-24% on MultiTP for models 3.8B and larger across 20 countries, without any weight updates.
LLMs exhibit identity-dependent hedging on human rights questions, with group identity as the strongest predictor among tested factors, and group steering mitigates the disparity.
AI value alignment is reconceptualized as a pluralistic governance problem arising along three axes—objectives, information, and principals—making it inherently context-dependent and unsolvable by technical design alone.
Case studies with blind UK residents and people from Kerala and Tamil Nadu demonstrate that community input at the systematization stage produces culturally grounded definitions of appropriateness for text-to-image model outputs.
citing papers explorer
-
ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
-
PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments
PERMA is a new benchmark using temporally ordered events, text variability, and linguistic alignment to evaluate LLM memory agents on persona consistency beyond simple retrieval.
-
From tools to thieves: Measuring and understanding public perceptions of AI through crowdsourced metaphors
Crowdsourced metaphors show rising anthropomorphism and warmth toward AI that predict trust and adoption, with notable demographic differences.
-
Retroactive Advantage Correction: Closed-Form V-Trace Bias Correction for Delay-Aware RLHF
RAC is a closed-form bias correction for delayed rewards in RLHF that is unbiased under full mass reinjection of the delay kernel and reduces to V-trace with no delay.
-
PEBS: Per-rater Empirical-Bayes Shrinkage for RLHF Reward-Model Calibration
PEBS applies Morris-James-Stein empirical-Bayes shrinkage to per-rater affine calibrators in RLHF, cutting within-user held-out RMSE by 8.58% on PRISM and 9.66% on PluriHarms versus pooled baselines.
-
Training-Free Cultural Alignment of Large Language Models via Persona Disagreement
DISCA converts within-country disagreement among World Values Survey personas into a bounded logit correction that reduces cultural misalignment by 10-24% on MultiTP for models 3.8B and larger across 20 countries, without any weight updates.
-
Hedging and Non-Affirmation: Quantifying LLM Alignment on Questions of Human Rights
LLMs exhibit identity-dependent hedging on human rights questions, with group identity as the strongest predictor among tested factors, and group steering mitigates the disparity.
-
Relative Principals, Pluralistic Alignment, and the Structural Value Alignment Problem
AI value alignment is reconceptualized as a pluralistic governance problem arising along three axes—objectives, information, and principals—making it inherently context-dependent and unsolvable by technical design alone.
-
Evaluating AI-Generated Images of Cultural Artifacts with Community-Informed Rubrics
Case studies with blind UK residents and people from Kerala and Tamil Nadu demonstrate that community input at the systematization stage produces culturally grounded definitions of appropriateness for text-to-image model outputs.
- Reinforcement Learning from Human Feedback