Fixed block causal masks create reachability boundaries where representations depend only on block prefixes, formalized via dependency sets and phase-conditioned coverage functions, with a parameter-free boundary bridge repair.
∞Bench: Extending long context evaluation beyond 100K tokens
8 Pith papers cite this work, alongside 14 external citations. Polarity classification is still indexing.
8
Pith papers citing it
14
external citations · OpenAlex
years
2026 8representative citing papers
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
EntropyInfer adaptively allocates inference compute using per-head attention entropy for rigid/dynamic classification during prefilling and compresses KV cache with generated tokens, achieving up to 2.39x speedup on long contexts.
IPO-Mine releases a toolkit and large multimodal dataset for structured analysis of IPO filings and shows state-of-the-art models diverge from human judgments on chart quality and misleadingness.
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
PRR accelerates dynamic sparse attention decoding in long-context LLMs via EMA-based prediction, speculative attention, and FlashAttention repair, achieving up to 40% latency reduction.
MOSAIC uses an Integer Linear Program scheduler for expert placement and prompt assignment plus adaptive aggregation to achieve 1.7-2.3x end-to-end speedup on 4-GPU MoA workloads while keeping accuracy within 0.1pp.
ATLAS is a length-dependent benchmarking framework that evaluates 26 models on 8 capability dimensions and shows substantial rank changes when moving from 128K to 1M token ranges.
citing papers explorer
-
Locality Does Not Imply Reachability: Boundary Repair in Block-Sparse Causal Attention
Fixed block causal masks create reachability boundaries where representations depend only on block prefixes, formalized via dependency sets and phase-conditioned coverage functions, with a parameter-free boundary bridge repair.
-
SCOUT: Active Information Foraging for Long-Text Understanding with Decoupled Epistemic States
SCOUT achieves state-of-the-art long-text understanding with up to 8x lower token use by actively foraging for sparse query-relevant information and updating a compact provenance-grounded epistemic state.
-
From Rigid to Dynamic: Entropy-Guided Adaptive Inference for Long-Context LLMs
EntropyInfer adaptively allocates inference compute using per-head attention entropy for rigid/dynamic classification during prefilling and compresses KV cache with generated tokens, achieving up to 2.39x speedup on long contexts.
-
IPO-Mine: A Toolkit and Dataset for Section-Structured Analysis of Long, Multimodal IPO Documents
IPO-Mine releases a toolkit and large multimodal dataset for structured analysis of IPO filings and shows state-of-the-art models diverge from human judgments on chart quality and misleadingness.
-
Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
-
Predict, Reuse, and Repair: Accelerating Dynamic Sparse Attention for Long-Context LLM Decoding
PRR accelerates dynamic sparse attention decoding in long-context LLMs via EMA-based prediction, speculative attention, and FlashAttention repair, achieving up to 40% latency reduction.
-
MOSAIC: Efficient Mixture-of-Agent Scheduling via Adaptive Aggregation and Inference Concurrency
MOSAIC uses an Integer Linear Program scheduler for expert placement and prompt assignment plus adaptive aggregation to achieve 1.7-2.3x end-to-end speedup on 4-GPU MoA workloads while keeping accuracy within 0.1pp.
-
ATLAS: All-round Testing of Long-context Abilities across Scales
ATLAS is a length-dependent benchmarking framework that evaluates 26 models on 8 capability dimensions and shows substantial rank changes when moving from 128K to 1M token ranges.