AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Differentiable Conformal Training for LLM Reasoning Factuality

Nathan Hittesdorf, Marco Salzetta, Lu Cheng

ICML 2026 · Large language models › Language, safety and interpretability · approach: Uncertainty quantification for LLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace hand-crafted frequency scoring with learned differentiable conformal scorer (DCF)Coherent Factuality · MATH (alpha=0.03)retention: 0.73 -> 1.76Improved
learned DCF scorer vs CFCoherent Factuality · MATH (alpha=0.10)retention: 3.33 -> 5.99Improved
learned DCF scorer vs CF at low alphaCoherent Factuality · FELM (alpha=0.01)retention (%): 11.1 -> 17.9Improved
learned DCF scorer vs CF at mid-range alphaCoherent Factuality · FELM (alpha=0.08)retention (%): 61.3 -> 52.7Hurt
learned DCF scorer vs CF across alphaCoherent Factuality · FELMretention (%): +61.3 at 0.01, -14.0 at 0.08Mixed
add graph-aware learning to combine features instead of single feature nx_reachability (ablation)CF · MATH (alpha=0.06)retention: 2.87 -> 3.56Improved
combine features vs claim_index only (ablation)CF · MATH (alpha=0.06)retention: 1.86 -> 3.56Improved
combine features vs frequency only (ablation)CF · FELM (alpha=0.01)retention (mean claims): 0.51 -> 0.72Improved
combine features vs frequency only at alpha 0.08 (ablation)CF · FELM (alpha=0.08)retention (mean claims): 2.56 -> 2.11Hurt
add learning without graph structure (Boosted Independent) versus non-learned Independent (ablation)Independent Factuality · MATH/FELMretentionHurt
add graph structure to Independent Factuality (CF) (ablation)Independent Factuality · MATH/FELMretentionImproved
replace differentiable conformal training with independently trained XGBoost scorer (ablation)DCF · MATH/FELMretentionHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy