Pith. sign in

REVIEW 15 cited by

Efficient Attentions for Long Document Summarization

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2104.02112 v2 pith:O5RQ5RNL submitted 2021-04-05 cs.CL

classification cs.CL
keywords efficientmodelsattentionsdocumentevaluationexistingheposlong
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

The quadratic computational and memory complexities of large Transformers have limited their scalability for long document summarization. In this paper, we propose Hepos, a novel efficient encoder-decoder attention with head-wise positional strides to effectively pinpoint salient information from the source. We further conduct a systematic study of existing efficient self-attentions. Combined with Hepos, we are able to process ten times more tokens than existing models that use full attentions. For evaluation, we present a new dataset, GovReport, with significantly longer documents and summaries. Results show that our models produce significantly higher ROUGE scores than competitive comparisons, including new state-of-the-art results on PubMed. Human evaluation also shows that our models generate more informative summaries with fewer unfaithful errors.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 15 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. When Local Variance Optimality Is Not Enough: RoPE-Aligned Q/K Rotations for Dynamic 4-Bit Quantisation

    cs.LG 2026-08 conditional novelty 6.0 of 10

    Under dynamic W4A4KV4 quantisation, a head-shared RoPE-aligned pairwise rotation that exactly minimizes a pooled variance surrogate still yields higher perplexity than full-head Hadamard mixing in all evaluated comparisons.

  2. CompilerKV: Risk-Adaptive KV Compression via Offline Experience Compilation

    cs.LG 2026-02 unverdicted novelty 6.0 of 10

    Offline-learned head-reliability and risk-threshold tables make prefill-only KV compression recover about 97.7% of uncompressed LongBench accuracy at a 512-token-per-layer memory budget.

  3. EvolKV: Evolutionary KV Cache Compression for LLM Inference

    cs.LG 2025-09 conditional novelty 6.0 of 10

    CMA-ES search over per-layer KV cache budgets beats uniform and pyramidal compression heuristics on LongBench, NIAH, RULER, and GSM8K, and edges past the full cache on one code dataset at 1.5% of the budget.

  4. Alignment-Augmented Speculative Decoding with Alignment Sampling and Conditional Verification

    cs.CL 2025-05 conditional novelty 6.0 of 10

    AASD combines alignment sampling from the prefilling distribution with entropy-based conditional verification to improve both the accuracy and speed of retrieval-based speculative decoding.

  5. CriticalKV: Optimizing KV Cache Eviction from an Output Perturbation Perspective

    cs.CL 2025-02 conditional novelty 6.0 of 10

    A new selection metric that combines attention weights with projected value-state norms reduces output perturbation in LLM key-value cache eviction.

  6. Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads

    cs.CL 2025-01 conditional novelty 6.0 of 10

    Task-KV identifies 'heterogeneous' attention heads by distance from a per-task semantic center and allocates differentiated KV cache budgets, achieving modest average gains on long-context benchmarks at reduced memory.

  7. AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference

    cs.CL 2025-01 conditional novelty 6.0 of 10

    AdaSkip speeds up long-context LLM inference by adaptively skipping low-importance attention and FFN sublayers in both the prompt-reading and token-generation phases, with quality tradeoffs.

  8. CaseSumm: A Large-Scale Dataset for Long-Context Summarization from U.S. Supreme Court Opinions

    cs.CL 2024-12 conditional novelty 6.0 of 10

    CaseSumm, a 25.6K-pair dataset of Supreme Court opinions and official syllabuses, shows automated metrics favor fine-tuned Mistral while human experts prefer GPT-4, and LLM judges do not align with humans better than ROUGE.

  9. RED-PIM: Reducing Data Movement for Transformers using Processing-in-Memory

    cs.LG 2026-07 conditional novelty 5.0 of 10

    RED-PIM restructures attention to use a compact d×d intermediate matrix, reducing PIM inter-bank data movement from O(N²) to O(N) and claiming up to 99.99% latency reduction in simulation.

  10. Fourier Position Embedding: Enhancing Attention's Periodic Extension for Length Generalization

    cs.AI 2024-12 conditional novelty 5.0 of 10

    FoPE replaces RoPE's single-frequency rotation per dimension with a Fourier series and clips under-trained low frequencies, improving length generalization in transformer LMs.

  11. Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models

    cs.CL 2024-12 conditional novelty 5.0 of 10

    Giving each attention head its own RoPE base frequency lets a single continual-pretraining stage at 128k match or beat a three-stage schedule, per the paper's NiaH, PPL and RULER results.

  12. From Global to Local: Learning Context-Aware Graph Representations for Document Classification and Summarization

    cs.CL 2026-02 conditional novelty 4.0 of 10

    Replacing full attention with sliding-window attention when building document sentence-graphs gives competitive classification accuracy at lower attention cost, with unfiltered local graphs best on medium-length news.

  13. HoPE: Hyperbolic Rotary Positional Encoding for Stable Long-Range Dependency Modeling in Large Language Models

    cs.CL 2025-09 reject novelty 4.0 of 10

    HoPE replaces RoPE's sine/cosine rotations with hyperbolic functions plus an exponential damping term to enforce monotonic attention decay, but the claimed consistent superiority and the 'RoPE as special case' theorem...

  14. PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference

    cs.LG 2025-09 conditional novelty 4.0 of 10

    PagedEviction prunes the KV cache in whole blocks using a key-value norm ratio, speeding up long-context LLM inference in vLLM while staying close to full-cache accuracy.

  15. More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache Compression

    cs.CL 2024-12 conditional novelty 4.0 of 10

    Storing 4x as many KV cache tokens at 4-bit precision outperforms storing 1x tokens at 16-bit precision on long-context benchmarks at equal data-memory budgets.

Pith tools