Invisible Unicode perturbations, optimized from surrogate compressors then adapted by prior-guided evolution under a low query budget, cause large information loss in agent context compression without changing human-visible text.
In: Findings of the Association for Computational Linguistics: ACL 2024 (Aug 2024)
12 Pith papers cite this work, alongside 32 external citations. Polarity classification is still indexing.
12
Pith papers citing it
32
external citations · OpenAlex
citation-role summary
background 3
citation-polarity summary
years
2026 12roles
background 3representative citing papers
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.
ClawVM introduces a harness-managed virtual memory system for LLM agents that ensures deterministic residency and durability of state under token budgets by using typed pages and validated writeback.
Ranking latent-memory fragments by middle-layer attention density and keeping only the top few more than doubles long-term knowledge-retention accuracy (43.0% vs. 17.4%/17.6%) versus MemoryLLM and M+.
SKIM is an adaptive multi-resolution soft-token framework that compresses procedural skills while aiming to preserve logical dependencies and task performance better than prior compression methods.
Cost-aware skill rewriting that preserves task-relevant operational anchors reduces LLM-agent total token cost by ~7% on a 20-task held-out panel and ~15% across agent stacks while maintaining verifier quality.
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.
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.
CoreMem replaces cosine retrieval with Fisher-Rao Riemannian matching and introduces Fisher-guided discrete token distillation for syntax-aware compression, reporting +4.51 pp open-domain and +4.17 pp temporal gains on LOCOMO and LongMemEval-S while staying inside an 8 GB VRAM budget.
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
-
Out of Sight: Compression-Aware Content Protection against Agentic Crawlers
Invisible Unicode perturbations, optimized from surrogate compressors then adapted by prior-guided evolution under a low query budget, cause large information loss in agent context compression without changing human-visible text.
-
TSCG: Deterministic Tool-Schema Compilation for Agentic LLM Deployments
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 Unified Model and Document Representation for On-Device Retrieval-Augmented Generation
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.
-
ClawVM: Harness-Managed Virtual Memory for Stateful Tool-Using LLM Agents
ClawVM introduces a harness-managed virtual memory system for LLM agents that ensures deterministic residency and durability of state under token budgets by using typed pages and validated writeback.
-
MemDefrag: Latent Memory Defragmentation for Large Language Models
Ranking latent-memory fragments by middle-layer attention density and keeping only the top few more than doubles long-term knowledge-retention accuracy (43.0% vs. 17.4%/17.6%) versus MemoryLLM and M+.
-
Adaptive Multi-Resolution Procedural Knowledge Compression for Large Language Models
SKIM is an adaptive multi-resolution soft-token framework that compresses procedural skills while aiming to preserve logical dependencies and task performance better than prior compression methods.
-
What Should a Skill Remember? Quality--Cost Trade-offs in Cost-Aware Skill Rewriting for Language Model Agents
Cost-aware skill rewriting that preserves task-relevant operational anchors reduces LLM-agent total token cost by ~7% on a 20-task held-out panel and ~15% across agent stacks while maintaining verifier quality.
-
CRAFT: Cost-aware Refinement And Front-aware Tuning of Prompts
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.
-
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.
-
CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents
CoreMem replaces cosine retrieval with Fisher-Rao Riemannian matching and introduces Fisher-guided discrete token distillation for syntax-aware compression, reporting +4.51 pp open-domain and +4.17 pp temporal gains on LOCOMO and LongMemEval-S while staying inside an 8 GB VRAM budget.
-
GRC: Unifying Reasoning-Driven Generation, Retrieval and Compression
GRC unifies generation, retrieval, and compression in LLMs via meta latent tokens for single-pass execution with modular flexibility.
-
LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
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.