SRCH:2A87D263
Block-Sparse FlashAttention Throughput Scaling on LongBench for Llama-3-70B at 256K Contexts
Abstract
Abstract: This report synthesises findings from 12 peer-reviewed papers addressing the following research question: How does Block-Sparse FlashAttention's throughput scale with increasing batch sizes compared to FlashAttention-2 on the LongBench benchmark for Llama-3-70b with 256k-token contexts. 9 claims were extracted from source literature; 9 were independently verified against retrieved documents. An automated multi-reviewer quality assessment produced a score of 5.0/10. This report is a machine-generated literature synthesis and does not constitute original research.
Research Question
How does Block-Sparse FlashAttention's throughput scale with increasing batch sizes compared to FlashAttention-2 on the LongBench benchmark for Llama-3-70b with 256k-token contexts?
Verification Level
| Paper level | L2, Source-grounded claims | |
| Source-grounded claims | 9 | |
| Claim record source | not publicly specified |
Descriptive public verification status only; aggregate claim counts are public, but individual claim records are not exposed here.
Quality Tier
| Tier | Watchlist | |
| Basis | Review score or public verified-claim signal is below DOI-grade threshold. |
Descriptive public triage only; this tier does not alter current publication or DOI behavior.
Quality Dimensions
| Evidence strength | LOW | |
| Citation grounding | MEDIUM | |
| Uncertainty disclosure | MEDIUM | |
| Reproducibility status | MEDIUM |
Automated triage signals derived from public fields; not human peer review or independent validation.
Correction Record
| Status | CURRENT |
| Correction count | 0 |
| Manifest contract | paper-manifest-v1.1 |
| Correction contract | correction-record-v1 |
Public corrections are additive records. Current status does not claim the synthesis is error-free.
Provenance
| Publisher | Assignee Research |
| Public provenance | L3, Claim aggregate record |
| Report artifact | Available |
| External record | Not registered |
| Claim lineage | 9 aggregate source-grounded claims |
| Review method | Automated multi-reviewer assessment |
| Quality guide | How to read scores, claims, manifests, and evidence links |
| Provenance contract | source-provenance-v1 |
| Note | Machine-generated synthesis of existing literature. Not primary research. |