AloLab, an iterative meta-agent prompt optimizer, raises structured output accuracy for 7-9B models from 0% to 84-87% on GSM8K while preserving near-native inference speed.
arXiv preprint arXiv:2403.01632 (2024)
8 Pith papers cite this work, alongside 3 external citations. Polarity classification is still indexing.
8
Pith papers citing it
3
external citations · external index
representative citing papers
TreeCoder improves LLM code generation accuracy by representing decoding as an optimizable tree search over programs with first-class constraints for syntax, style, and execution, outperforming baselines on MBPP and SQL-Spider.
RefineStat improves small language model performance on probabilistic program synthesis by adding semantic constraint enforcement and diagnostic-aware refinement, producing syntactically and statistically reliable code that often matches larger models.
Empirical benchmarks on four SE tasks show grammar-constrained decoding and TTMG eliminate most syntax errors in LLM outputs while structural and semantic errors persist and cascade in downstream tools.
TruncProof is a grammar-constrained decoding method that enforces token-length limits on LLM-generated JSON by using LL(1) parser lookahead to estimate remaining tokens needed for syntactic validity.
AdaDec improves Pass@1 accuracy of LLM code generation by up to 20.9% over greedy decoding by triggering lookahead reranking only at high-uncertainty steps on HumanEval+, MBPP+, and DevEval.
Compiled AI generates deterministic code artifacts from LLMs in a one-time compilation step, enabling reliable workflow execution with zero runtime tokens after break-even.
Fine-tuning codegen-mono on 500 synthetic-plus-reviewed Python examples produces 99% accuracy, 98.08% precision, 100% recall and 99.04% F1 for CWE detection.
citing papers explorer
-
When Correct Isn't Usable: Improving Structured Output Reliability in Small Language Models
AloLab, an iterative meta-agent prompt optimizer, raises structured output accuracy for 7-9B models from 0% to 84-87% on GSM8K while preserving near-native inference speed.
-
TreeCoder: Systematic Exploration and Optimisation of Decoding and Constraints for LLM Code Generation
TreeCoder improves LLM code generation accuracy by representing decoding as an optimizable tree search over programs with first-class constraints for syntax, style, and execution, outperforming baselines on MBPP and SQL-Spider.
-
RefineStat: Efficient Exploration for Probabilistic Program Synthesis
RefineStat improves small language model performance on probabilistic program synthesis by adding semantic constraint enforcement and diagnostic-aware refinement, producing syntactically and statistically reliable code that often matches larger models.
-
Empirical Study for Structured Output Control in LLMs for Software Engineering
Empirical benchmarks on four SE tasks show grammar-constrained decoding and TTMG eliminate most syntax errors in LLM outputs while structural and semantic errors persist and cascade in downstream tools.
-
TruncProof: A Guardrail for LLM-based JSON Generation under Token-Length Constraints
TruncProof is a grammar-constrained decoding method that enforces token-length limits on LLM-generated JSON by using LL(1) parser lookahead to estimate remaining tokens needed for syntactic validity.
-
AdaDec: A Uncertainty-Guided Lookahead Decoding Framework for LLM-Based Code Generation
AdaDec improves Pass@1 accuracy of LLM code generation by up to 20.9% over greedy decoding by triggering lookahead reranking only at high-uncertainty steps on HumanEval+, MBPP+, and DevEval.
-
Compiled AI: Deterministic Code Generation for LLM-Based Workflow Automation
Compiled AI generates deterministic code artifacts from LLMs in a one-time compilation step, enabling reliable workflow execution with zero runtime tokens after break-even.
-
Case Study: Fine-tuning Small Language Models for Accurate and Private CWE Detection in Python Code
Fine-tuning codegen-mono on 500 synthetic-plus-reviewed Python examples produces 99% accuracy, 98.08% precision, 100% recall and 99.04% F1 for CWE detection.