Belief-Trajectory Energy: Measuring the Path to a Prediction
cs.CL
Submitted: 2026-10-04
Updated: 2026-10-04
Project page: https://yingjiahao14.github.io/BTE-web
Sources
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks
- Model Utility Law: Evaluating LLMs beyond Performance through Mechanism Interpretable Metric
- The Llama 3 Herd of Models
- GPT-4o System Card
- Language Models (Mostly) Know What They Know
- Let's Verify Step by Step
- 2 OLMo 2 Furious
- OpenAI GPT-5 System Card
- Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
- EffiEval: Efficient and Generalizable Model Evaluation via Capability Coverage Maximization
- Beyond Benchmarks: Understanding Mixture-of-Experts Models through Internal Mechanisms
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering