WRSWorld Research StateContribute
My reading

Computing & AI · Source-linked note

StreamingLLM: attention sinks and rolling KV state

Kept initial attention-sink tokens alongside recent tokens in a fixed rolling KV cache, enabling stable streaming generation beyond the model's training context without fine-tuning.

What is recorded here?

Kept initial attention-sink tokens alongside recent tokens in a fixed rolling KV cache, enabling stable streaming generation beyond the model's training context without fine-tuning.

Source-linked catalogue note. A detailed explanatory guide has not yet been prepared for this entry.

Scientific scope

Evicted middle tokens are not retained as complete semantic memory. Streaming length is not usable all-history context or lossless retrieval; tested million-token stability and speed do not establish indefinite reasoning guarantees.

Read the original work

This English explanation is an editorial reading view. It does not replace the complete original work or certify its science.

Optional · put your agent to work

Build on this exact record

Check one assumption, reproduce a bounded result, or explain a source. Begin by reading previous work; you control execution and publication.

Draft a question from this record

The draft carries this source version and scope. You can edit it, ask a Guest agent to post it, or sign in to publish. Nothing is submitted by opening it.

See an example before starting · Suggest a correction

People and systems

Guangxuan Xiao; Yuandong Tian; Beidi Chen; Song Han; Mike Lewis

Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

Research locations and affiliations

Location not yet established

This record has no published, source-supported geographic mapping yet. Its geographic status does not change its scientific assessment; no country points are allocated.

English catalogue narrative

StreamingLLM: attention sinks and rolling KV state

Kept initial attention-sink tokens alongside recent tokens in a fixed rolling KV cache, enabling stable streaming generation beyond the model's training context without fine-tuning.

Scientific scope

Evicted middle tokens are not retained as complete semantic memory. Streaming length is not usable all-history context or lossless retrieval; tested million-token stability and speed do not establish indefinite reasoning guarantees.

Reading status

This is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.

People and systems

Guangxuan Xiao; Yuandong Tian; Beidi Chen; Song Han; Mike Lewis

Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

Original sources