# Parser Transitions Already Reveal Structural Responsibility: KV Persistence for Structured Generation

arXiv: https://arxiv.org/abs/2608.28276 · HTML: https://mlsanity.com/paper/2608.28276/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 34 references that are in the map.
- First posted on arXiv 28 Aug 2026.

## Builds on (most cited first, 27 of 27)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs](https://mlsanity.com/paper/2305.03111/) (2023)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [Scissorhands: Exploiting the Persistence of Importance Hypothesis for LLM KV Cache Compression at Test Time](https://mlsanity.com/paper/2305.17118/) (2023)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [Guiding LLMs The Right Way: Fast, Non-Invasive Constrained Generation](https://mlsanity.com/paper/2403.06988/) (2024)
- [ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2502.00299/) (2025)
- [DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs](https://mlsanity.com/paper/2412.14838/) (2024)
- [RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression](https://mlsanity.com/paper/2502.14051/) (2025)
- [Fast KV Compaction via Attention Matching](https://mlsanity.com/paper/2602.16284/) (2026)
- [IterGen: Iterative Semantic-aware Structured LLM Generation with Backtracking](https://mlsanity.com/paper/2410.07295/) (2024)
- [CriticalKV: Optimizing KV Cache Eviction from an Output Perturbation Perspective](https://mlsanity.com/paper/2502.03805/) (2025)
- [SCOPE: Optimizing Key-Value Cache Compression in Long-context Generation](https://mlsanity.com/paper/2412.13649/) (2024)
- [FastKV: Decoupling of Context Reduction and KV Cache Compression for Prefill-Decoding Acceleration](https://mlsanity.com/paper/2502.01068/) (2025)
- [XGrammar-2: Dynamic and Efficient Structured Generation Engine for Agentic LLMs](https://mlsanity.com/paper/2601.04426/) (2026)
- [Don't Waste Bits! Adaptive KV-Cache Quantization for Lightweight On-Device LLMs](https://mlsanity.com/paper/2604.04722/) (2026)
- [IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference](https://mlsanity.com/paper/2606.09916/) (2026)
- [ArborKV: Structure-Aware KV Cache Management for Scaling Tree-based LLM Reasoning](https://mlsanity.com/paper/2605.22106/) (2026)
- [StructKV: Preserving the Structural Skeleton for Scalable Long-Context Inference](https://mlsanity.com/paper/2604.06746/) (2026)
- [TriAxialKV: Toward Extreme Low-Precision KV-Cache Quantization for Agentic Inference Tasks](https://mlsanity.com/paper/2605.17170/) (2026)
- [CodeComp: Structural KV Cache Compression for Agentic Coding](https://mlsanity.com/paper/2604.10235/) (2026)
- [Transactional Attention: Semantic Sponsorship for KV-Cache Retention](https://mlsanity.com/paper/2604.11288/) (2026)
- [Adaptive Filtering of the KV Cache: Diagnosing and Correcting Structural-Role Bias in LLM Inference](https://mlsanity.com/paper/2607.13205/) (2026)

## Abstract (arXiv)

LLM agents use structured generation for JSON, SQL, and function calls. Constrained decoding ensures valid formats, but KV compression can discard history needed for correct function and argument selection. Structural responsibility can diverge from model saliency: importance scores can overlook generated tokens' structural roles. Parser transitions already computed for legality encode these roles, including required arguments and structural boundaries, providing a free retention signal. We introduce PASK (Parser-Aware Structural KV Persistence), which reuses this signal for KV persistence by layer group without online importance estimation. Task-error sensitivity sets protection floors, while attention-output distortion ranks upgrades under the remaining budget. Offline calibration produces a frozen table for structural lookup and budget accounting. PASK outperforms budget-controlled baselines on function calling and text-to-SQL, improving BFCL accuracy by up to 8.67 pp over the strongest comparator. Against Full KV, it cuts peak GPU memory by 47.5 percent and achieves up to 3.3 times TPOT speedup.