Lesser Wrong Feed

 A J-Space-Based Metric for Model Valence: Defining the Metric, Testing, and Comparisons to Self-Reports
 Fri, 21 Aug 2026 21:54:35 GMT
 Alignment fine-tuning induces conditional misalignment in Qwen2.5-7B-Instruct
 Fri, 21 Aug 2026 20:48:04 GMT
 When is Unlimited Optimization Catastrophic?
 Fri, 21 Aug 2026 20:08:18 GMT
 AI Text Watermarking Is Free And Good
 Fri, 21 Aug 2026 19:30:08 GMT
 Evaluating Explanations of LLM Behavior In The Wild with Counterfactual Experiments
 Fri, 21 Aug 2026 19:09:21 GMT
 In Defense of ASI Socialism
 Fri, 21 Aug 2026 18:40:28 GMT
 Misaligned AI in the Bronze Age
 Fri, 21 Aug 2026 16:51:55 GMT
 The imposters among us: function vectors that ace every check and do the wrong task (in search of circularity)
 Fri, 21 Aug 2026 15:58:35 GMT
 When Models Identify as a Swarm
 Fri, 21 Aug 2026 11:49:37 GMT
 My Neel Nanda MATS 10.0 Application: Studying Feature Splitting in SAEs via Training Data Attribution
 Fri, 21 Aug 2026 09:41:09 GMT