From Neural Network Decisions to Training Cases: An Exact Account via Case-Based Decision Theory

• URL: http://arxiv.org/abs/2607.11347v1
• Date: Mon, 13 Jul 2026 10:08:00 GMT
• Title: From Neural Network Decisions to Training Cases: An Exact Account via Case-Based Decision Theory
• Abstract summary: We show that an OLS action readout fitted on a fixed neural representation admits an exact case-based decomposition.<n>The decomposition yields audit signals that trace scores to training cases, measure action coherence, and identify weak support.
• Score: 2.394494660847436
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Neural networks increasingly guide decisions in high-stakes domains such as medical diagnosis, credit approval, and energy bidding. Audit in these settings requires case-level evidence: which training cases support an action and what outcomes they carried. Case-based decision theory (CBDT) formalizes this reasoning by aggregating outcome support from remembered cases. We show that an OLS action readout fitted on a fixed neural representation admits an exact case-based decomposition. Each action score is a weighted sum of training-case returns, with coefficients determined by empirical Gram geometry. We identify a sufficient regime for CBDT similarity semantics; outside it, the coefficients should generally be treated as signed Gram-geometric influence. The decomposition yields audit signals that trace scores to training cases, measure action coherence, and identify weak support. Across synthetic CBDT, PJM, Adult Income, and Default Credit tasks, the method recovers case-level preference structure and achieves the highest mean Top-30 consistency among compared attribution baselines, while remaining competitive on support reconstruction. The audit requires only fitting an OLS top-layer probe, without retraining the representation or accessing the original optimization trajectory; probe fidelity is measured by score reconstruction.

Related papers

• Gated Decoupled Compositional Bandits: A Unified Theory of Contextual Bandits with Supervised-Calibrated Action Scaling and Pre-Execution Gating [0.0]
  Gated Decoupled Compositional Bandits (GDCB)<n>GDCB is a family of contextual bandit algorithms with three structural innovations.<n>We formalise this class of algorithms, prove four structural theorems characterising its statistical behaviour.
  arXiv  Detail & Related papers  (2026-08-22T14:58:25Z)
• More Data, Worse Decisions? Preference Reversals in Neural Networks under Gram Incompatibility [9.196757973061198]
  Case-Based Decision Theory (CBDT) formalizes the requirement through its composition axiom.<n>We study when this property holds for fixed-representation neural networks with ordinary least squares (OLS) output heads.<n>We develop a three-stage audit that traces strict pairwise reversals through decision changes to task-defined utility loss.
  arXiv  Detail & Related papers  (2026-07-28T18:52:47Z)
• Beyond Perplexity: A Behavioral Evaluation Framework for Deployment-Memory Claims in LLM Test-Time Training [30.22002300048916]
  Large language model test-time training (TTT) is often evaluated through local proxy metrics.<n>We introduce a behavioral evaluation framework that calibrates TTT memory claims to the evidence that supports them.
  arXiv  Detail & Related papers  (2026-07-01T03:07:17Z)
• Process-Verified Reinforcement Learning for Theorem Proving via Lean [47.294932573003756]
  We show that the Lean proof assistant itself can serve as a symbolic process oracle.<n>Experiments with LeanLean and DeepSeek-V1.5 show that tactic-level supervision outperforms outcome-only baselines in most settings.
  arXiv  Detail & Related papers  (2026-06-18T10:40:55Z)
• BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation [51.22170603236523]
  We propose a Balanced Token-level Policy Optimization framework for hallucination mitigation.<n>BalTO extracts checkable factual claims, verifies them against the reference context, and projects claim-level judgments to token-level labels.<n>Experiments on ConFiQA, RAGTruth, and FinLLM-Eval show that BALTO achieves the highest faithfulness across all six model--benchmark settings.
  arXiv  Detail & Related papers  (2026-06-14T16:25:59Z)
• RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards [76.17893114021757]
  Deep research systems that plan, search, evaluate evidence, and synthesize long-form reports lack ground-truth answers and span many tool-augmented decisions.<n>In this work, we argue that rubrics should serve not merely as final-answer evaluators, but as the shared interface that structures policy execution, judge feedback, and agent memory.<n>We introduceEM, a rubric-guided reinforcement learning framework that combines stagewise policy decomposition with reflection-based meta-policy evolution.
  arXiv  Detail & Related papers  (2026-05-11T17:40:38Z)
• Reasoning Provenance for Autonomous AI Agents: Structured Behavioral Analytics Beyond State Checkpoints and Execution Traces [0.0]
  Agent Execution Record (AER) is a structured reasoning primitive that captures intent, observation, and inference as first-class queryable fields on every step.<n>We show how AERs enable population-level behavioral analytics: reasoning pattern mining, confidence calibration, cross-agent comparison, counterfactual regression testing via mock replay.
  arXiv  Detail & Related papers  (2026-03-23T08:27:54Z)
• The Emergence of Lab-Driven Alignment Signatures: A Psychometric Framework for Auditing Latent Bias and Compounding Risk in Generative AI [0.0]
  This paper introduces a novel auditing framework to quantify latent trait estimation under ordinal uncertainty.<n>The research audits nine leading models across dimensions including Optimization Bias, Sycophancy, and Status-Quo Legitimization.
  arXiv  Detail & Related papers  (2026-02-19T06:56:01Z)
• APR: Penalizing Structural Redundancy in Large Reasoning Models via Anchor-based Process Rewards [61.52322047892064]
  Test-Time Scaling (TTS) has significantly enhanced the capabilities of Large Reasoning Models (LRMs)<n>We observe that LRMs frequently conduct repetitive self-verification without revision even after obtaining the final answer during the reasoning process.<n>We propose Anchor-based Process Reward (APR), a structure-aware reward shaping method that localizes the reasoning anchor and penalizes exclusively the post-anchor AST.
  arXiv  Detail & Related papers  (2026-01-31T14:53:20Z)
• Beyond Outcome Verification: Verifiable Process Reward Models for Structured Reasoning [14.632557283678898]
  Verifiable Process Reward Models (VPRMs) are a reinforcement-learning framework in which intermediate reasoning steps are checked by deterministic, rule-based verifiers.<n>We apply VPRMs to risk-of-bias assessment for medical evidence synthesis.<n>Results show that VPRMs achieve up to 20% higher F1 than state-of-the-art models and 6.5% higher than verifiable outcome rewards.
  arXiv  Detail & Related papers  (2026-01-23T23:22:20Z)
• Gaming the Judge: Unfaithful Chain-of-Thought Can Undermine Agent Evaluation [76.5533899503582]
  Large language models (LLMs) are increasingly used as judges to evaluate agent performance.<n>We show this paradigm implicitly assumes that the agent's chain-of-thought (CoT) reasoning faithfully reflects both its internal reasoning and the underlying environment state.<n>We demonstrate that manipulated reasoning alone can inflate false positive rates of state-of-the-art VLM judges by up to 90% across 800 trajectories spanning diverse web tasks.
  arXiv  Detail & Related papers  (2026-01-21T06:07:43Z)
• Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoning [52.144281362465996]
  We propose EAPO (Evidence-Augmented Policy Optimization) to apply Reinforcement Learning to long-context scenarios.<n>We first establish the Evidence-Augmented Reasoning paradigm, validating via Tree-Structured Evidence Sampling.<n>We then introduce a specialized RL algorithm where a reward model computes a Group-Relative Evidence Reward.<n>To sustain accurate supervision throughout training, we further incorporate an Adaptive Reward-Policy Co-Evolution mechanism.
  arXiv  Detail & Related papers  (2026-01-15T11:40:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.