Zstandard
PulseAugur coverage of Zstandard — every cluster mentioning Zstandard across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Cadence uses foundation model for error-bounded time series compression
Researchers have developed Cadence, a novel error-bounded lossy compression system for time series data. Cadence integrates Google's TimesFM-3 foundation model with an adaptive arithmetic coder to ensure sample-level ac…
-
Loreley system uses Quality-Diversity search for large-scale program evolution
Researchers have developed Loreley, a new system designed for large-scale program evolution using Quality-Diversity (QD) search. Unlike traditional methods that either accumulate changes from a single best version or re…
-
Diffusion LMs advance lossless text compression, beating LLMs and zstd
Researchers have introduced Diffusion Language Models (DLMs) as a novel approach to lossless text compression, aiming to overcome the throughput limitations of existing autoregressive LLM-based methods. This new framewo…
-
Simon Willison prototypes SQLite text history compression
Simon Willison has proposed a novel method for storing revision histories in SQLite databases by compressing a JSON array of all previous text versions. This approach leverages compression algorithms like zlib or Zstand…
-
AI agents could benefit from new token-native storage method
A new research paper proposes "token-native storage" as a more efficient method for storing text data used by AI agents. Instead of traditional UTF-8 encoding, the paper suggests storing text directly as token IDs, whic…
-
Brevis system synthesizes programs for lossless tensor compression
Researchers have developed Brevis, a novel system that treats lossless tensor compression as a program synthesis problem. Brevis utilizes a domain-specific language (DSL) with reversible operators to capture tensor stru…
-
Gamers compress 80TB Minecraft server map to 15TB using custom format · 2 sources tracked
A team of Minecraft players has successfully mapped and compressed the 80TB data of the game's oldest anarchy server, 2b2t, down to a 15TB download. This monumental preservation effort, named 2b2tplace, took 22 months a…
-
LLMs achieve 82% better source code compression than general tools
Researchers have developed a new method for lossless source code compression using Large Language Models (LLMs) that improves upon existing techniques. This approach, called Thresholded Symbol Ranking, bounds LLM predic…