JevCascade: Readout Stability and Inference-Time Compute Allocation for Prefill-Only Decision Models
cs.CL
Submitted: 2026-10-06
Updated: 2026-10-10
Code: https://github.com/rlisml/jev-cascade
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
- Reasoning Models Don't Always Say What They Think
- this-that-model-1.0: A typed decision model that decides in 30 ms, for a millionth of a cent
- Learning to Route LLMs with Confidence Tokens
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces
- A Unified Approach to Routing and Cascading for LLMs
- Evaluating and Benchmarking the System One Model Jev
- Language Model Cascades
- Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures
- Measuring Massive Multitask Language Understanding
- Language Models (Mostly) Know What They Know
- Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents
- Localize-Then-Decide Guarantees for LLM Judgments
- Overwhelmed by Choice: Studying LLM Decision Making at Scale
- Explaining CLIP Zero-shot Predictions Through Concepts
- Bridging the Knowledge-Prediction Gap in LLMs on Multiple-Choice Questions
- Reasoning Models are Test Exploiters: Rethinking Multiple-Choice
- Fine-Tuning Large Language Models to Appropriately Abstain with Semantic Entropy
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