From Statute to Control Flow: Span-Grounded Deontic Trees for Defeasible Scope Parsing

• URL: http://arxiv.org/abs/2606.08932v1
• Date: Mon, 08 Jun 2026 02:17:53 GMT
• Title: From Statute to Control Flow: Span-Grounded Deontic Trees for Defeasible Scope Parsing
• Abstract summary: Rule-following agents often fail via Silent Scope Omission (SSO)<n>The underlying bottleneck is statutory and policy understanding.<n>NormBench is a benchmark of 2,290 provisions spanning Chinese (laws and local policies), English (U.S. tax law), and corporate policies.
• Score: 13.077273684682886
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Rule-following agents tasked with executing policies and regulations often fail via Silent Scope Omission (SSO): a model applies a general rule but silently drops nested exceptions or counter-exceptions, producing outputs that appear compliant yet break on important edge cases. Although such failures are often framed as an agentic-systems problem, the underlying bottleneck is statutory and policy understanding, a capability typically studied in legal NLP. However, most existing legal NLP benchmarks emphasize end-task outcomes, which can overlook the structural omissions that cause SSO. To diagnose and mitigate SSO, we introduce NormBench, a benchmark of 2,290 provisions spanning Chinese (laws and local policies), English (U.S. tax law, GDPR, and corporate policies), and cross-lingual settings, designed for defeasible scope parsing: identifying precisely which clause overrides which. NormBench uses Span-Grounded Deontic Trees (SG-DT), a compiler-style intermediate representation that anchors every logical branch to source spans and requires explicit exclusion guards, enabling deterministic compilation and audit. Evaluations of frontier LLMs reveal two recurring pathologies: (1) Recursion Decay, where performance drops sharply as defeater depth increases, and (2) an Auditability Trap, where models retrieve relevant spans but fail to assemble correct control flow. Using SG-DT as a constrained intermediate output improves whole-tree fidelity and defeater recovery, and downstream experiments show that its utility is mechanism-specific: gains concentrate on exception-active, SSO-prone cases, while aggregate accuracy can be mixed when the added structure is unnecessary or parser fidelity is low.

Related papers

• Grounded Normative Rule Generation with Structured Search [47.28595196084396]
  We formalize the problem as Grounded Normative Rule Synthesis (GNRS)<n>We introduce GNRS-Search, a framework that utilizes Markov Chain Monte Carlo (MCMC) sampling to optimize a discrete, five-slot And-Or Graph (AOG)<n>We evaluate our approach on GNRS-Bench, a benchmark spanning 116 controlled goals across eight scene families, and RealCharterBench, which evaluates transfer to 53 real-derived policy tasks with hidden source clauses.
  arXiv  Detail & Related papers  (2026-08-23T05:43:42Z)
• Citation-Closure Retrieval and Per-Rule Attribution for Real-World Regulatory Compliance Question Answering [54.97384993676565]
  We formalize Regulatory Compliance QA with RegOps-Bench, a novel benchmark featuring an Operational Knowledge Graph derived from complex national R&D regulations.<n>RefWalk traverses cross-document citations, fuses multi-view candidates via max-based aggregation, and enforces per-rule attribution to explicitly map claims to sources.
  arXiv  Detail & Related papers  (2026-05-28T10:38:38Z)
• Interactive Critique-Revision Training for Reliable Structured LLM Generation [18.00222080273147]
  We propose DPA-GRPO, a paired-action training method for a two-player generator--verifier game with structured verifier interventions.<n>We analyze the unregularized game and show that positive probability on strictly lower-reward intervention or revision actions creates a profitable unilateral deviation.<n>Experiments on TaxCalc TY24 show that DPA-GRPO improves structured decision accuracy over zero-shot generation and generator-only RL baselines.
  arXiv  Detail & Related papers  (2026-05-08T17:00:38Z)
• Logic-Regularized Verifier Elicits Reasoning from LLMs [63.65875399266337]
  We propose LOVER, an unsupervised verifier regularized by logical rules.<n>Lover treats theverifier as a binary latent variable, utilizinginternal activations and enforcing three logical constraints.<n>Lover significantly outperforms unsupervised baselines.
  arXiv  Detail & Related papers  (2026-05-07T09:03:49Z)
• Accurate Legal Reasoning at Scale: Neuro-Symbolic Offloading and Structural Auditability for Robust Legal Adjudication [0.0]
  Legal texts often contain computational legal clauses, whose understanding requires complex logic.<n>We propose Amortized Intelligence, a neuro-symbolic approach where we use an LLM once to translate a legal text into Deterministic Autonomous Contract Language.<n>Our DACL-based Agent achieves near-perfect consistency and mitigates the "reasoning cliff" observed in probabilistic models.
  arXiv  Detail & Related papers  (2026-05-04T11:13:59Z)
• Job Skill Extraction via LLM-Centric Multi-Module Framework [51.37063712967151]
  We present SRICL, an LLM-centric framework that combines semantic retrieval, in-context learning, and supervised fine-tuning.<n>On six public span-labeled corpora of job-ad sentences, SRICL achieves substantial STRICT-F1 improvements over GPT-3.5 prompting baselines and sharply reduces invalid tags and hallucinated spans.
  arXiv  Detail & Related papers  (2026-04-23T10:46:07Z)
• PSR2: A Phase-based Semantic Reasoning Framework for Atomicity Violation Detection via Contract Refinement [9.36682660657815]
  PSRtextsuperscript2 is a novel collaborative static analysis framework that integrates structural path searching with deterministic semantic reasoning.<n>Experiments on 1,600 contract samples demonstrate that PSRtextsuperscript2 significantly outperforms pattern-matching baselines.<n>Our fusion logic effectively reduces the false-positive rate by nearly half compared to single module analysis.
  arXiv  Detail & Related papers  (2026-04-08T11:46:01Z)
• Improving Search Agent with One Line of Code [68.58667107354253]
  Tool-based Agentic Reinforcement Learning (TARL) has emerged as a promising paradigm for training search agents to interact with external tools.<n>We propose textbfSearch textbfAgent textbfPolicy textbfOptimization (textbfSAPO), which stabilizes training via a conditional token-level KL constraint.
  arXiv  Detail & Related papers  (2026-03-10T04:07:39Z)
• Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models [57.20761595019967]
  We present Normalized Attention Guidance (NAG), an efficient, training-free mechanism that applies extrapolation in attention space with L1-based normalization and refinement.<n>NAG restores effective negative guidance where CFG collapses while maintaining fidelity.<n>NAG generalizes across architectures (UNet, DiT), sampling regimes (few-step, multi-step), and modalities (image, video)
  arXiv  Detail & Related papers  (2025-05-27T13:30:46Z)

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.