{"total":24,"rows":[{"id":"2510.04340","title":"Inoculation Prompting: Eliciting traits from LLMs during training can suppress them at test-time","created":"2025-10-05","cited_by_count":43,"area":null},{"id":"2605.02087","title":"Model Spec Midtraining: Improving How Alignment Training Generalizes","created":"2026-05-03","cited_by_count":20,"area":null},{"id":"2601.21571","title":"Shaping capabilities with token-level data filtering","created":"2026-01-29","cited_by_count":13,"area":null},{"id":"2509.22367","title":"What Is The Political Content in LLMs' Pre- and Post-Training Data?","created":"2025-09-26","cited_by_count":7,"area":null},{"id":"2509.23383","title":"Train Once, Answer All: Many Pretraining Experiments for the Cost of One","created":"2025-09-27","cited_by_count":6,"area":null},{"id":"2510.02978","title":"AI Generated Child Sexual Abuse Material -- What's the Harm?","created":"2025-10-03","cited_by_count":5,"area":null},{"id":"2605.12705","title":"Early Data Exposure Improves Robustness to Subsequent Fine-Tuning","created":"2026-05-12","cited_by_count":4,"area":null},{"id":"2606.06533","title":"Position: Don't Just \"Fix it in Post\": A Science of AI Must Study Training Dynamics","created":"2026-06-03","cited_by_count":3,"area":null},{"id":"2606.11660","title":"Bergson: An Open Source Library for Data Attribution","created":"2026-06-10","cited_by_count":3,"area":null},{"id":"2607.08077","title":"Modular Pretraining Enables Access Control","created":"2026-07-09","cited_by_count":3,"area":null},{"id":"2602.05164","title":"Position: Capability Control Should be a Separate Goal From Alignment","created":"2026-02-05","cited_by_count":2,"area":null},{"id":"2605.06901","title":"Reflections and New Directions for Human-Centered Large Language Models","created":"2026-05-07","cited_by_count":2,"area":null},{"id":"2510.27629","title":"Best Practices for Biorisk Evaluations on Open-Weight Bio-Foundation Models","created":"2025-10-31","cited_by_count":1,"area":null},{"id":"2601.01584","title":"Steerability of Instrumental-Convergence Tendencies in LLMs","created":"2026-01-04","cited_by_count":1,"area":null},{"id":"2601.23157","title":"No More, No Less: Least-Privilege Language Models","created":"2026-01-30","cited_by_count":1,"area":null},{"id":"2608.13482","title":"Synthetic Persona Pretraining: Alignment from Token Zero","created":"2026-08-13","cited_by_count":1,"area":null},{"id":"2609.37914","title":"The Unequal Influence of Bad Advice: Using Training Data Attribution to Modulate Emergent Misalignment","created":"2026-09-29","cited_by_count":1,"area":null},{"id":"2602.02258","title":"Alignment-Aware Model Adaptation via Feedback-Guided Optimization","created":"2026-02-02","cited_by_count":0,"area":null},{"id":"2605.28843","title":"The Biosecurity Blind Spot: Systematic Dual-use Detection in Open Science Infrastructure","created":"2026-05-10","cited_by_count":0,"area":null},{"id":"2606.05029","title":"Validity Threats for Foundation Model Research","created":"2026-06-03","cited_by_count":0,"area":null},{"id":"2607.05407","title":"Position: Preventing AI-Generated CSAM Necessitates New Approaches to AI Safety","created":"2026-06-09","cited_by_count":0,"area":null},{"id":"2609.00519","title":"The Safeguard Worked. Is the LLM System Safer?","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.20027","title":"Can Data Attribution Filter Out Subliminal Learning? Not Reliably","created":"2026-09-17","cited_by_count":0,"area":null},{"id":"2610.09004","title":"Removing Information Content Does Not Certify Tamper Resistance in Open-Weight Models","created":"2026-10-06","cited_by_count":0,"area":null}],"next_cursor":null}