LatentSkill uses a hypernetwork to generate LoRA adapters from textual skills, enabling weight-space storage that cuts prefill tokens and boosts agent success rates on ALFWorld and Search-QA.
hub Canonical reference
LLMLingua: Com- pressing prompts for accelerated inference of large language models
Canonical reference. 80% of citing Pith papers cite this work as background.
17
Pith papers citing it
Background
80% of classified citations
hub tools
citation-role summary
background 4
baseline 1
citation-polarity summary
representative citing papers
Mobius Injection exploits semantic closure in LLM agents to enable single-message AbO-DDoS attacks achieving up to 51x call amplification and 229x latency inflation.
TSCG compiles JSON tool schemas into token-efficient structured text, raising tool-use accuracy for small LLMs from 0% to 84.4% on benchmarks while cutting tokens by 52-57%.
A single model unifies retrieval and context compression for on-device RAG via shared representations, matching traditional RAG performance at 1/10 context size with no extra storage.
KV cache compression causes task-dependent degradation in high-density reasoning due to disrupted CoT links; ShotKV mitigates this by preserving few-shot examples as indivisible semantic units through phase separation, delivering 9-18% accuracy gains and 11% latency reduction.
LongMemEval benchmarks long-term memory in chat assistants, revealing 30% accuracy drops across sustained interactions and proposing indexing-retrieval-reading optimizations that boost performance.
RouterBench supplies a standardized benchmark, 405k+ inference dataset, theoretical framework, and comparative analysis for multi-LLM routing systems.
Input compression increases net cost and reduces accuracy as models respond longer; output compression reduces cost on most models, with surface text diverging from unconstrained references on non-reasoning models.
Dropout-GRPO uses structured dropout to generate trajectory variance for GRPO in latent-reasoning models like Coconut, raising GSM8K pass@1 from 27.29% to 29.01%.
CRAFT is a Pareto-front prompt optimizer that allocates scarce LLM validation calls to candidates near the current front using accuracy- and cost-oriented generators plus NSGA-II retention.
Combining local routing with prompt compression saves 45-79% cloud tokens on edit and explanation workloads, while a fuller set including draft-review saves 51% on RAG-heavy tasks.
FACT-E uses controlled perturbations as an instrumental signal to measure intra-chain faithfulness in CoT reasoning and combines it with answer consistency to select trustworthy trajectories.
LightThinker++ adds explicit adaptive memory management and a trajectory synthesis pipeline to LLM reasoning, cutting peak token use by ~70% while gaining accuracy in standard and long-horizon agent tasks.
LLMLingua prompt compression yields up to 18% end-to-end LLM speedups with unchanged quality when prompt length, ratio, and hardware align, plus an open profiler to predict the break-even point.
Empirical study demonstrates that cost-aware skill rewriting for LLM agents can achieve 7% total cost reduction and 6% agent-token cost reduction with preserved quality on SkillsBench.
GRC unifies generation, retrieval, and compression in LLMs via meta latent tokens for single-pass execution with modular flexibility.
LensVLM trains VLMs to scan compressed rendered text images and selectively expand task-relevant regions, achieving 4.3x compression with near full-text accuracy and outperforming baselines up to 10.1x on text QA benchmarks.
citing papers explorer
-
Prompt Compression in the Wild: Measuring Latency, Rate Adherence, and Quality for Faster LLM Inference
LLMLingua prompt compression yields up to 18% end-to-end LLM speedups with unchanged quality when prompt length, ratio, and hardware align, plus an open profiler to predict the break-even point.