Pith. sign in

REVIEW 7 cited by

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 2410.23771 v5 pith:HD4RXF72 submitted 2024-10-31 cs.CL cs.LG

What is Wrong with Perplexity for Long-context Language Modeling?

classification cs.CL cs.LG
keywords long-contexttokensllmslongpplacrossbenchmarkscapabilitiescontext
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Handling long-context inputs is crucial for large language models (LLMs) in tasks such as extended conversations, document summarization, and many-shot in-context learning. While recent approaches have extended the context windows of LLMs and employed perplexity (PPL) as a standard evaluation metric, PPL has proven unreliable for assessing long-context capabilities. The underlying cause of this limitation has remained unclear. In this work, we provide a comprehensive explanation for this issue. We find that PPL overlooks key tokens, which are essential for long-context understanding, by averaging across all tokens and thereby obscuring the true performance of models in long-context scenarios. To address this, we propose \textbf{LongPPL}, a novel metric that focuses on key tokens by employing a long-short context contrastive method to identify them. Our experiments demonstrate that LongPPL strongly correlates with performance on various long-context benchmarks (e.g., Pearson correlation of -0.96), significantly outperforming traditional PPL in predictive accuracy. Additionally, we introduce \textbf{LongCE} (Long-context Cross-Entropy) loss, a re-weighting strategy for fine-tuning that prioritizes key tokens, leading to consistent improvements across diverse benchmarks. In summary, these contributions offer deeper insights into the limitations of PPL and present effective solutions for accurately evaluating and enhancing the long-context capabilities of LLMs. Code is available at https://github.com/PKU-ML/LongPPL.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

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

  1. How Data Shapes RoPE Frequency Usage: From Positional Scale Matching to Length Generalization

    cs.LG 2026-07 conditional novelty 7.0

    RoPE frequency usage is determined by a data-induced dependency width W, with the optimal frequency scaling as π/W, explaining both learned spectra and the success of position interpolation.

  2. Pretraining Recurrent Networks without Recurrence

    cs.LG 2026-06 unverdicted novelty 6.0

    SMT reduces RNN training to supervised learning on memory transitions (m_t, x_{t+1}) to m_{t+1} obtained from a Transformer encoder, enabling time-parallel training with O(1) gradient paths.

  3. Where Does Long-Context Supervision Actually Go? Effective-Context Exposure Balancing

    cs.CL 2026-05 conditional novelty 6.0

    EXACT re-allocates training supervision by inverse frequency of long effective-context targets, improving NoLiMa and RULER scores by 5-18 points on Qwen and LLaMA models without degrading standard QA or reasoning.

  4. On the Importance and Evaluation of Narrativity in Natural Language AI Explanations

    cs.CL 2026-04 unverdicted novelty 6.0

    XAI explanations should be narratives with continuous structure, cause-effect, fluency and diversity, and new metrics are needed to evaluate this better than standard NLP scores.

  5. PolicyLong: Towards On-Policy Context Extension

    cs.LG 2026-04 unverdicted novelty 6.0

    PolicyLong shifts long-context data synthesis to an on-policy loop that re-screens contexts using the evolving model's entropy landscape, producing a self-curriculum that outperforms static offline baselines with larg...

  6. Understanding the Performance Plateau in Text-to-Video Retrieval: A Comprehensive Empirical and Linguistic Analysis

    cs.IR 2026-03 unverdicted novelty 6.0

    Short, simple captions describing single actions achieve higher retrieval recall than complex multi-step or fine-grained scene descriptions across all tested models.

  7. Gated KalmaNet: A Fading Memory Layer Through Test-Time Ridge Regression

    cs.LG 2025-11 unverdicted novelty 6.0

    Gated KalmaNet uses exact Kalman gain computation with adaptive gating and Chebyshev iteration to improve SSM performance on long-context tasks over prior approximations like DeltaNet.