Trafy
Research

What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness

Large language models fine-tuned for forecasting can be accurate yet poorly calibrated, and their chain-of-thought (CoT) reasoning may not faithfully reflect...

Pratyush Ranjan Tiwari·Jul 9, 2026·2 min read·Original source ↗
What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness

What LLM Forecasters Know but Don't Say: Probing Internal Representations for Calibration and Faithfulness2607.08046AuthorsPratyush Ranjan Tiwari,Siddharth Boppana,Christopher J. Earls,Srikar Varadaraj,Eric Hoand 1 moreAbstractLarge language models fine-tuned for forecasting can be accurate yet poorly calibrated, and their chain-of-thought (CoT) reasoning may not faithfully reflect the evidence behind a forecast. We ask whether internal representations offer a more direct window into both. Working with Eternis-Forecaster 8B on OpenForesight, we train representation-pooling probes on intermediate activations and find they achieve substantially better calibration; a result that also holds for GLM-4.7-Flash and GLM-4.5-Air. We then assess CoT faithfulness through evidence ablation and diversionary injection: removing an influential source in the prompt often changes the model's forecast while leaving the reasoning trace untouched. The same probes function as lie detectors: their activations track behavioral shifts far better than the reasoning trace does, and they also predict the direction of change in 84% of cases, including when the CoT conceals the perturbation's influence. Finally, forced answering reveals that forecasts are largely fixed before reasoning begins: a single pre-reasoning pass recovers the committed answer and confidence, and routing questions by the spread of this pre-set answer distribution saves 30-47% of generated tokens, with no loss of accuracy. Together, these results establish probing internal representations as a practical tool for calibrating, auditing, and triaging language model forecasters and reasoning models more broadly.ResourcesView on Hugging FaceRead PDFArXiv

Related

Learning When to Trust via Selective Context Preference OptimizationResearch

Learning When to Trust via Selective Context Preference Optimization

Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong. The obvious remedy, training models to resist such signals, hides a failure mode: a model that ignores all context looks robust yet is useless when the context is worth trusting. We recast the problem as selective trust and introduce MIST, a human-annotated benchmark that renders each reasoning item under four matched conditions (clean, misleading, correct-context, and irrelevant-context), together with SC2W, a paired metric counting how often a misle

arXiv (cs.AI) · Aug 6, 2026
4 min
Learning When to Trust via Selective Context Preference OptimizationResearch

Learning When to Trust via Selective Context Preference Optimization

Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong. The obvious remedy, tr...

Papers with Code · Aug 6, 2026
1 min
Tracing the Heart: An Evidence-Linked Pipeline for Heart-Failure Feature EngineeringResearch

Tracing the Heart: An Evidence-Linked Pipeline for Heart-Failure Feature Engineering

Electronic health record (EHR) feature engineering is a major bottleneck in clinical research and AI, accounting for 39-45% of data scientists' workload. This is especially pronounced in heart failure, which affects an estimated 6.7 million U.S. adults and requires integrating fragmented EHR data with disease-specific, guideline-based clinical reasoning. Existing rule-based and large language model (LLM)-based approaches offer only partial automation with limited maintainability and evidence traceability. We developed the Nimblemind Multi-Agent System (nMAS), an evidence-linked, rubr

arXiv (cs.AI) · Aug 6, 2026
4 min