Worst-Case Symbolic Constraints Analysis and Generalisation with Large Language Models

• URL: http://arxiv.org/abs/2506.08171v2
• Date: Tue, 16 Sep 2025 10:35:33 GMT
• Title: Worst-Case Symbolic Constraints Analysis and Generalisation with Large Language Models
• Authors: Daniel Koh, Yannic Noller, Corina S. Pasareanu, Adrians Skapars, Youcheng Sun,
• Abstract summary: Worst-case symbolic constraints analysis requires inferring the symbolic constraints that characterise worst-case program executions.<n>We show that even state-of-the-art large language models (LLMs) struggle when applied directly on this task.<n>We propose WARP, an innovative neurosymbolic approach that computes worst-case constraints on smaller concrete input sizes.
• Score: 7.658134651527103
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models (LLMs) have demonstrated strong performance on coding tasks such as generation, completion and repair, but their ability to handle complex symbolic reasoning over code still remains underexplored. We introduce the task of worst-case symbolic constraints analysis, which requires inferring the symbolic constraints that characterise worst-case program executions; these constraints can be solved to obtain inputs that expose performance bottlenecks or denial-of-service vulnerabilities in software systems. We show that even state-of-the-art LLMs (e.g., GPT-5) struggle when applied directly on this task. To address this challenge, we propose WARP, an innovative neurosymbolic approach that computes worst-case constraints on smaller concrete input sizes using existing program analysis tools, and then leverages LLMs to generalise these constraints to larger input sizes. Concretely, WARP comprises: (1) an incremental strategy for LLM-based worst-case reasoning, (2) a solver-aligned neurosymbolic framework that integrates reinforcement learning with SMT (Satisfiability Modulo Theories) solving, and (3) a curated dataset of symbolic constraints. Experimental results show that WARP consistently improves performance on worst-case constraint reasoning. Leveraging the curated constraint dataset, we use reinforcement learning to fine-tune a model, WARP-1.0-3B, which significantly outperforms size-matched and even larger baselines. These results demonstrate that incremental constraint reasoning enhances LLMs' ability to handle symbolic reasoning and highlight the potential for deeper integration between neural learning and formal methods in rigorous program analysis.

Related papers

• Rethinking On-Device LLM Reasoning: Why Analogical Mapping Outperforms Abstract Thinking for IoT DDoS Detection [8.874341026403854]
  This paper introduces a novel detection framework that integrates Chain-of-Thought (CoT) reasoning with Retrieval-Augmented Generation (RAG)<n>We systematically evaluate compact ODLLMs, including LLaMA 3.2 (1B, 3B) and Gemma 3 (1B, 4B), using structured prompting and exemplar-driven reasoning strategies.<n> Experimental results demonstrate substantial performance improvements with few-shot prompting, achieving macro-average F1 scores as high as 0.85.
  arXiv  Detail & Related papers  (2026-01-20T15:18:56Z)
• LLM-Guided Quantified SMT Solving over Uninterpreted Functions [10.767268261124515]
  Quantified formulas with Uninterpreted Functions (UFs) over non-linear real arithmetic pose fundamental challenges for Satisfiability Modulo Theories (SMT) solving.<n>We present AquaForte, a framework that leverages Large Language Models to provide semantic guidance for UF instantiation.<n>Our approach preprocesses formulas through constraint separation, uses structured prompts to extract mathematical reasoning from LLMs, and integrates the results with traditional SMT algorithms.
  arXiv  Detail & Related papers  (2026-01-08T07:40:37Z)
• When Do Symbolic Solvers Enhance Reasoning in Large Language Models? [9.02964480989444]
  Large Reasoning Models (LRMs) achieve strong performance on complex reasoning tasks by generating long Chains of Thought (CoTs)<n>A promising direction is the symbolic-solver-integrated approach, which leverages the code generation capabilities of LLMs to translate reasoning tasks into executable code and then solve them with a symbolic solver.<n>Our experimental results show that the symbolic-solver-integrated method only helps when the problem requires limited implicit reasoning but involves an ample search space.
  arXiv  Detail & Related papers  (2025-12-02T22:23:26Z)
• Unlocking Symbol-Level Precoding Efficiency Through Tensor Equivariant Neural Network [84.22115118596741]
  We propose an end-to-end deep learning (DL) framework with low inference complexity for symbol-level precoding.<n>We show that the proposed framework captures substantial performance gains of optimal SLP, while achieving an approximately 80-times speedup over conventional methods.
  arXiv  Detail & Related papers  (2025-10-02T15:15:50Z)
• Discrete Tokenization for Multimodal LLMs: A Comprehensive Survey [69.45421620616486]
  This work presents the first structured taxonomy and analysis of discrete tokenization methods designed for large language models (LLMs)<n>We categorize 8 representative VQ variants that span classical and modern paradigms and analyze their algorithmic principles, training dynamics, and integration challenges with LLM pipelines.<n>We identify key challenges including codebook collapse, unstable gradient estimation, and modality-specific encoding constraints.
  arXiv  Detail & Related papers  (2025-07-21T10:52:14Z)
• Do LLMs Dream of Discrete Algorithms? [0.7646713951724011]
  Large Language Models (LLMs) have rapidly transformed the landscape of artificial intelligence.<n>Their reliance on probabilistic inference limits their effectiveness in domains requiring strict logical reasoning.<n>This paper proposes a neurosymbolic approach that augments LLMs with logic-based reasoning modules.
  arXiv  Detail & Related papers  (2025-06-29T22:03:01Z)
• A Theory of Inference Compute Scaling: Reasoning through Directed Stochastic Skill Search [15.387256204743407]
  Large language models (LLMs) demand considerable computational, energy, and financial resources during both training and deployment.<n>Inference costs now represent a significant and growing component of the overall resource burden.<n>We introduce directed skill search (DS3), a general framework that represents inference as expressive over a learned skill graph.
  arXiv  Detail & Related papers  (2025-06-10T14:47:48Z)
• LogicPuzzleRL: Cultivating Robust Mathematical Reasoning in LLMs via Reinforcement Learning [29.047063129464494]
  Large language models (LLMs) excel at many supervised tasks but often struggle with structured reasoning unfamiliar settings.<n>This discrepancy suggests that standard fine-tuning pipelines may instill narrow, domain-specifics rather than fostering general-purpose thinking strategies.<n>We propose a "play to learn" framework that fine-tunes LLMs through reinforcement learning on a suite of seven custom logic puzzles.
  arXiv  Detail & Related papers  (2025-06-05T09:40:47Z)
• Computational Thinking Reasoning in Large Language Models [69.28428524878885]
  Computational Thinking Model (CTM) is a novel framework that incorporates computational thinking paradigms into large language models (LLMs)<n>Live code execution is seamlessly integrated into the reasoning process, allowing CTM to think by computing.<n>CTM outperforms conventional reasoning models and tool-augmented baselines in terms of accuracy, interpretability, and generalizability.
  arXiv  Detail & Related papers  (2025-06-03T09:11:15Z)
• Scalable Chain of Thoughts via Elastic Reasoning [61.75753924952059]
  Elastic Reasoning is a novel framework for scalable chain of thoughts.<n>It separates reasoning into two phases--thinking and solution--with independently allocated budgets.<n>Our approach produces more concise and efficient reasoning even in unconstrained settings.
  arXiv  Detail & Related papers  (2025-05-08T15:01:06Z)
• CrossWordBench: Evaluating the Reasoning Capabilities of LLMs and LVLMs with Controllable Puzzle Generation [53.452699232071495]
  CrossWordBench is a benchmark designed to evaluate the reasoning capabilities of Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) through the medium of crossword puzzles.<n>Our evaluation reveals that reasoning LLMs outperform non-reasoning models substantially by effectively leveraging crossing-letter constraints.<n>Our findings offer insights into the limitations of the reasoning capabilities of current LLMs and LVLMs, and provide an effective approach for creating multimodal constrained tasks for future evaluations.
  arXiv  Detail & Related papers  (2025-03-30T20:03:36Z)
• DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs [86.76714527437383]
  This paper proposes DSMoE, a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks.<n>We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge.<n>Experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches.
  arXiv  Detail & Related papers  (2025-02-18T02:37:26Z)
• ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning [92.76959707441954]
  We introduce ZebraLogic, a comprehensive evaluation framework for assessing LLM reasoning performance.<n>ZebraLogic enables the generation of puzzles with controllable and quantifiable complexity.<n>Our results reveal a significant decline in accuracy as problem complexity grows.
  arXiv  Detail & Related papers  (2025-02-03T06:44:49Z)
• Attribute Controlled Fine-tuning for Large Language Models: A Case Study on Detoxification [76.14641982122696]
  We propose a constraint learning schema for fine-tuning Large Language Models (LLMs) with attribute control. We show that our approach leads to an LLM that produces fewer inappropriate responses while achieving competitive performance on benchmarks and a toxicity detection task.
  arXiv  Detail & Related papers  (2024-10-07T23:38:58Z)
• Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis [9.865771016218549]
  This work presents a novel systematic methodology to analyse the capabilities and limitations of Large Language Models (LLMs)<n>The analysis is complexity-graded w.r.t. rule dependency structure, allowing quantification of specific inference challenges on LLM performance.
  arXiv  Detail & Related papers  (2024-08-15T16:41:00Z)
• Q*: Improving Multi-step Reasoning for LLMs with Deliberative Planning [53.6472920229013]
  Large Language Models (LLMs) have demonstrated impressive capability in many natural language tasks. LLMs are prone to produce errors, hallucinations and inconsistent statements when performing multi-step reasoning. We introduce Q*, a framework for guiding LLMs decoding process with deliberative planning.
  arXiv  Detail & Related papers  (2024-06-20T13:08:09Z)
• Investigating Symbolic Capabilities of Large Language Models [16.88906206735967]
  This study aims to bridge the gap by rigorously evaluating Large Language Models (LLMs) on a series of symbolic tasks. Our analysis encompasses eight LLMs, including four enterprise-grade and four open-source models, of which three have been pre-trained on mathematical tasks. The findings reveal a significant decline in LLMs' performance on context-free and context-sensitive symbolic tasks as the complexity, represented by the number of symbols, increases.
  arXiv  Detail & Related papers  (2024-05-21T21:24:34Z)
• Characterization of Large Language Model Development in the Datacenter [55.9909258342639]
  Large Language Models (LLMs) have presented impressive performance across several transformative tasks. However, it is non-trivial to efficiently utilize large-scale cluster resources to develop LLMs. We present an in-depth characterization study of a six-month LLM development workload trace collected from our GPU datacenter Acme.
  arXiv  Detail & Related papers  (2024-03-12T13:31:14Z)
• DiLA: Enhancing LLM Tool Learning with Differential Logic Layer [11.810200077863172]
  We propose a novel differential logic layer-aided language modeling (DiLA) approach, where logical constraints are integrated into the forward and backward passes of a network layer. We evaluate the performance of DiLA on two classic reasoning problems and empirically demonstrate its consistent outperformance against existing prompt-based and solver-aided approaches.
  arXiv  Detail & Related papers  (2024-02-19T07:38:57Z)
• A General Framework for Learning from Weak Supervision [93.89870459388185]
  This paper introduces a general framework for learning from weak supervision (GLWS) with a novel algorithm. Central to GLWS is an Expectation-Maximization (EM) formulation, adeptly accommodating various weak supervision sources. We also present an advanced algorithm that significantly simplifies the EM computational demands.
  arXiv  Detail & Related papers  (2024-02-02T21:48:50Z)
• Evaluating LLMs' Mathematical and Coding Competency through Ontology-guided Interventions [47.83142414018448]
  We focus on two popular reasoning tasks: arithmetic reasoning and code generation. We introduce (i) a general ontology of perturbations for math and coding questions, (ii) a semi-automatic method to apply these perturbations, and (iii) two datasets. We show a significant performance drop across all the models against perturbed questions.
  arXiv  Detail & Related papers  (2024-01-17T18:13:07Z)

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.