mlsanityData to 14 Oct 2026

AMemOps: Evaluating Memory Updates in Long Audio Streams

arXiv 2610.18652 ↗First posted 13 Oct 2026cs.SD cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Audio memory must recognise recurring sources across acoustic changes while keeping their attributes up to date. Overall accuracy does not reveal how these demands interact. We introduce Audio Memory Operations (AMemOps), a benchmark for latest-attribute retrieval, preservation of unchanged attributes and absent-source rejection in long audio streams. Paired controls vary recordings, training identities and earlier events to examine their effects on updating. Training on other recordings of the test instruments increases a reference delta-rule memory's update advantage over matched attention by 13.4 percentage points. At retrieval delays of 120,s, matched histories separate benefits of consistent repetition from costs of conflicting earlier values, while additive-memory controls identify benefits of erasure.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 28 of its 31 references that are in the map. First posted on arXiv 13 Oct 2026.

Builds on

28 of 31 references.

References · 28

  1. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  2. 2020Transformers are RNNs: Fast Autoregressive Transformers with Linear Attentioncited by 2,022
  3. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Dualitycited by 1,336
  4. 2014Neural Turing Machinescited by 644
  5. 2024Gated Delta Networks: Improving Mamba2 with Delta Rulecited by 532
  6. 2023Gated Linear Attention Transformers with Hardware-Efficient Trainingcited by 514
  7. 2021Linear Transformers Are Secretly Fast Weight Programmerscited by 502
  8. 2024Parallelizing Linear Transformers with the Delta Rule over Sequence Lengthcited by 396
  9. 2020Query-Key Normalization for Transformerscited by 280
  10. 2017Neural Audio Synthesis of Musical Notes with WaveNet Autoencoderscited by 225
  11. 2023Zoology: Measuring and Improving Recall in Efficient Language Modelscited by 204
  12. 2025Kimi Linear: An Expressive, Efficient Attention Architecturecited by 196
16 more
  1. 2025Audio Flamingo 2: An Audio-Language Model with Long-Audio Understanding and Expert Reasoning Abilitiescited by 133
  2. 2024Mechanistic Design and Scaling of Hybrid Architecturescited by 59
  3. 2019Temporal Reasoning via Audio Question Answeringcited by 39
  4. 2026Gated DeltaNet-2: Decoupling Erase and Write in Linear Attentioncited by 33
  5. 2025AudioMarathon: A Comprehensive Benchmark for Long-Context Audio Understanding and Efficiency in Audio LLMscited by 22
  6. 2025Audio MultiChallenge: A Multi-Turn Evaluation of Spoken Dialogue Systems on Natural Human Interactioncited by 19
  7. 2025Mechanistic evaluation of Transformers and state space modelscited by 9
  8. 2026MemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversationscited by 7
  9. 2026Erase-then-Delta Attention: Decoupling Erase and Write Addresses in Delta-Rule Linear Attentioncited by 5
  10. 2021NAAQA: A Neural Architecture for Acoustic Question Answeringcited by 3
  11. 2026Kernelized Linear Attention: Breaking the Capacity Wall with Symmetric Conescited by 3
  12. 2026Event-Grounded Question Answering over Long Audio via Structured Retrievalcited by 2
  13. 2026Why Can't They Remember? Uncovering Representation and Retrieval Bottlenecks in Multi-Turn Acoustic Memorycited by 2
  14. 2026RAIL: Rethinking Auditory Intelligence in Large Audio-Language Models with a CHC-Grounded Benchmarkcited by 1
  15. 2026The Query Knows What to Forget: A Second Erase Direction for Linear Attentioncited by 1
  16. 2026Do SpeechLMs Hear Their Own Opinions? Diagnosing and Mitigating Previous-Belief Contamination in Streaming Emotion Understandingcited by 1

Built on by

Not cited yet.