Fermion Propagators: Stability of Self-Energy
29 Oct 2025
This section details the diagrammatic rules for computing the fermion propagator in Non-Fermi Liquids where interactions are mediated
Fermion Propagators: Bosonization vs. Holography
29 Oct 2025
This note addresses the challenge of computing fermion propagators in a broad class of strongly correlated systems
The Strength of Dynamic Encoding: RECKONING Outperforms Zero-Shot GPT-3.5 in Distractor Robustness
29 Oct 2025
RECKONING's performance significantly surpasses both zero-shot and few-shot GPT-3.5 prompting.
Ablation Study Confirms Necessity of Dynamic Rates for RECKONING Performance
29 Oct 2025
This article presents an ablation study confirming that an adaptive, per-step-per-layer learning rate is essential for the RECKONING framework.
Technical Setup for RECKONING: Inner Loop Gradient Steps, Learning Rates, and Hardware Specification
29 Oct 2025
This article outlines the implementation details for RECKONING, which uses a GPT-2-base model and runs on NVIDIA A100 GPUs.
Multi-Task vs. Single-Task ICR: Quantifying the High Sensitivity to Distractor Facts in Reasoning
29 Oct 2025
The results highlight ICR's vulnerability to interference and motivate the need for more robust, distraction-mitigating approaches like RECKONING.
Evaluating Systematic Generalization: The Use of ProofWriter and CLUTRR-SG in LLM Reasoning Research
28 Oct 2025
This article provides a detailed description of two multi-hop logical reasoning datasets: ProofWriter and CLUTRR-SG.
Meta-Learning for Reasoning: Summary of RECKONING's Superior Performance and Future Impact
28 Oct 2025
This conclusion summarizes RECKONING, a novel bi-level learning framework that robustly solves multi-hop reasoning problems
Connecting Reasoning, Model Editing, and LLM Knowledge: RECKONING's Contribution to AI Research
28 Oct 2025
This article situates RECKONING within three major research fields: Logical Reasoning, Model Editing, and Language Models as Knowledge Bases.