Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs | Paper Citation Record

Observation cf29194d-1cfa-489f-9ed6-5e7190f31635 · outbound

This paper cites Instruction Tuning with GPT-4.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Instruction Tuning with GPT-4

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.449922Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation eea683b8-7027-4e02-bdaf-74fad92d9296 · outbound

This paper cites The Llama 3 Herd of Models.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs The Llama 3 Herd of Models

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.423879Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c43ec401-fa32-4f5c-a5b2-44a87e99bc8d · outbound

This paper cites Qwen3 Technical Report.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Qwen3 Technical Report

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.409134Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 0d59d235-b32b-4ae5-8cbf-67d63b774509 · outbound

This paper cites Phi-4 Technical Report.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Phi-4 Technical Report

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.420712Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation bd8f82fb-f82f-4345-8fae-52185feaca23 · outbound

This paper cites SALMONN: Towards generic hearing abilities for large language models,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs SALMONN: Towards generic hearing abilities for large language models,

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.713988Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 0eb31cc9-519b-4462-a17f-166f730a67b2 · outbound

This paper cites Qwen2.5 Technical Report.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Qwen2.5 Technical Report

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.438595Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c9693dc2-8f6b-4e79-90e4-a082da1e3011 · outbound

This paper cites Alignformer: Modality matching can achieve better zero-shot instruction-following speech-llm,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Alignformer: Modality matching can achieve better zero-shot instruction-following speech-llm,

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.711851Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3dea55de-3b2e-4df7-8ec7-13b8c4fc7a5e · outbound

This paper cites On The Landscape of Spoken Language Models: A Comprehensive Survey.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs On The Landscape of Spoken Language Models: A Comprehensive Survey

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.411923Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation eab642b6-1e4e-4f1b-9f97-bc41e9d4ecab · outbound

This paper cites Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.433275Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation f01a2625-132d-4e4e-aa75-f02d060a22cf · outbound

This paper cites DeSTA: Enhancing speech language models through descriptive speech-text alignment,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs DeSTA: Enhancing speech language models through descriptive speech-text alignment,

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.718414Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 524c9804-a579-4ad3-95df-7c8496bf23dc · outbound

This paper cites Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning Data,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Developing Instruction-Following Speech Language Model Without Speech Instruction-Tuning Data,

Reference 11

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.705608Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 39cc6a20-8cbc-4dab-9d92-a69221744adf · outbound

This paper cites Desta2.5- audio: Toward general-purpose large audio language model with self- generated cross-modal alignment,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Desta2.5- audio: Toward general-purpose large audio language model with self- generated cross-modal alignment,

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.691218Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6db8e2bd-cf83-472c-9e4b-ace4cdc07a2b · outbound

This paper cites Prompting large language models with speech recognition abilities,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Prompting large language models with speech recognition abilities,

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.701557Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3fe8e6f2-f109-49ef-bf16-6fcddfd29a9d · outbound

This paper cites WavLLM: Towards Robust and Adaptive Speech Large Language Model.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs WavLLM: Towards Robust and Adaptive Speech Large Language Model

Reference 14

Resolution: metadata mismatch; local_arxiv, observed 2026-07-10T20:37:34.398583Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3f7e0a52-5d58-43fa-926d-e60fa96128a2 · outbound

This paper cites An Embarrassingly Simple Approach for LLM with Strong ASR Capacity.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs An Embarrassingly Simple Approach for LLM with Strong ASR Capacity

Reference 15

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.441426Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 083a14fa-dd81-48bd-b888-c00565990f3a · outbound

This paper cites Wav2Prompt: End-to-end speech prompt learning and task-based fine-tuning for text-based LLMs,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Wav2Prompt: End-to-end speech prompt learning and task-based fine-tuning for text-based LLMs,

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.699667Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 2e081245-57f6-4186-a128-ad1f9b815477 · outbound

This paper cites Connecting speech encoder and large language model for asr,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Connecting speech encoder and large language model for asr,

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.709409Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3354767c-0d74-4cf4-9ad5-3cdc80b03b4f · outbound

This paper cites Speech-xl: Towards long-form speech understanding in large speech language models,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Speech-xl: Towards long-form speech understanding in large speech language models,

Reference 18

Resolution: verified exact; arxiv_id, observed 2026-07-10T20:37:34.402206Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 93a7ca83-4585-4a38-be2e-f3f596bf8444 · outbound

This paper cites Speech llms are contextual reasoning transcribers,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Speech llms are contextual reasoning transcribers,

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-07-10T20:37:34.427244Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 7327d8b1-cd30-4a3f-a60d-6e21164a307c · outbound

This paper cites On decoder-only architecture for speech-to-text and large language model integration,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs On decoder-only architecture for speech-to-text and large language model integration,

Reference 20

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.729136Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6ac338bb-48c3-4918-b3fb-a060126c5936 · outbound

This paper cites Speechmap- per: Speech-to-text embedding projector for llms,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Speechmap- per: Speech-to-text embedding projector for llms,

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-07-10T20:37:34.417901Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b2da3027-61ab-4387-bd63-21bfc808a640 · outbound

This paper cites Snapkv: Llm knows what you are looking for before generation,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Snapkv: Llm knows what you are looking for before generation,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.752754Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 56f0e0a1-1c5a-4778-8880-38c2ef289fe3 · outbound

This paper cites Minicache: Kv cache compression in depth dimension for large language models,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Minicache: Kv cache compression in depth dimension for large language models,

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.744945Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 705beebf-fb9f-4fa9-9e07-f513173d625a · outbound

This paper cites Open asr leaderboard: Towards reproducible and transparent multilingual and long-form speech recognition evaluation,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Open asr leaderboard: Towards reproducible and transparent multilingual and long-form speech recognition evaluation,

Reference 24

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.754734Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation fb5ffe7d-8af3-4bed-96ea-283851fa926a · outbound

This paper cites Efficient memory management for large language model serving with pagedattention.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Efficient memory management for large language model serving with pagedattention

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.742944Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6151166a-3fd3-4b0b-bf07-b82ebaa178a9 · outbound

This paper cites Connection- ist temporal classification: labelling unsegmented sequence data with recurrent neural networks,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Connection- ist temporal classification: labelling unsegmented sequence data with recurrent neural networks,

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.693481Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e7e404c7-3a2f-4810-9a5c-6fe5a7be0dbe · outbound

This paper cites Cjst: Ctc compressor based joint speech and text training for decoder-only asr,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Cjst: Ctc compressor based joint speech and text training for decoder-only asr,

Reference 27

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.737191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 9ed1227e-f456-4693-b887-5afe4eafbb15 · outbound

This paper cites How to Connect Speech Foundation Models and Large Language Models? What Matters and What Does Not.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs How to Connect Speech Foundation Models and Large Language Models? What Matters and What Does Not

Reference 28

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.414645Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6f6605fe-fcbd-4554-836d-5b64e9564ff1 · outbound

This paper cites Cif: Continuous integrate-and-fire for end-to- end speech recognition,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Cif: Continuous integrate-and-fire for end-to- end speech recognition,

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.695666Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation aba02e2f-c556-4b4f-8ffc-e855301556bc · outbound

This paper cites BLIP-2: Bootstrapping language- image pre-training with frozen image encoders and large language mod- els,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs BLIP-2: Bootstrapping language- image pre-training with frozen image encoders and large language mod- els,

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.741068Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 14c54e88-8246-4c65-b344-c9b593196b10 · outbound

This paper cites Distilling an End-to-End Voice Assistant Without Instruction Training Data.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Distilling an End-to-End Voice Assistant Without Instruction Training Data

Reference 31

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.405190Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3c3ff200-163d-464e-aa09-ffc4b311661c · outbound

This paper cites A survey on large language model acceleration based on KV cache management,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs A survey on large language model acceleration based on KV cache management,

Reference 32

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.748991Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c2e7dfad-44dd-419f-bb26-d97f8db5b63c · outbound

This paper cites H2o: Heavy- hitter oracle for efficient generative inference of large language models,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs H2o: Heavy- hitter oracle for efficient generative inference of large language models,

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.747054Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 4113fdae-1bc2-41fb-aa09-0c8b85b1857e · outbound

This paper cites Efficient streaming language models with attention sinks,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Efficient streaming language models with attention sinks,

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.724774Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ee2be897-38d7-4d83-bcc2-c0a7c1c5875b · outbound

This paper cites Kvquant: Towards 10 million context length llm inference with kv cache quantization,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Kvquant: Towards 10 million context length llm inference with kv cache quantization,

Reference 35

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.726897Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 3602818e-f029-493f-a7e6-d2610c0c9e6b · outbound

This paper cites Dynamic memory compression: Retrofitting LLMs for accelerated inference,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Dynamic memory compression: Retrofitting LLMs for accelerated inference,

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.722786Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 0dcc80aa-866d-4b6f-8816-53c67589c206 · outbound

This paper cites Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks

Reference 37

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.430154Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation dc8867a3-fc92-451b-a903-ce4cb85dd9d9 · outbound

This paper cites Adapting language models to compress contexts,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Adapting language models to compress contexts,

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.720595Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b828db19-f4ab-4881-b9fc-778cb926758f · outbound

This paper cites MLS: A Large-Scale Multilingual Dataset for Speech Research.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs MLS: A Large-Scale Multilingual Dataset for Speech Research

Reference 39

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.436040Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 6969e927-329e-4f93-9d8b-f5e11c135e37 · outbound

This paper cites GigaSpeech: An Evolving, Multi-Domain ASR Corpus with 10,000 Hours of Transcribed Audio,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs GigaSpeech: An Evolving, Multi-Domain ASR Corpus with 10,000 Hours of Transcribed Audio,

Reference 40

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.731112Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a262608f-fd45-4870-b55b-a47a2fde477a · outbound

This paper cites SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition

Reference 41

Resolution: metadata mismatch; local_arxiv, observed 2026-07-10T20:37:34.447307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation e37a5331-43c5-4246-802e-547170de6731 · outbound

This paper cites Librispeech: an asr corpus based on public domain audio books.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Librispeech: an asr corpus based on public domain audio books

Reference 42

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.703497Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c15ce9cb-2b5e-4118-a982-ba21284cf367 · outbound

This paper cites Libriheavy: a 50,000 hours asr corpus with punctuation casing and context,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Libriheavy: a 50,000 hours asr corpus with punctuation casing and context,

Reference 43

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.707462Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 399d939d-6696-48df-8dd6-5391eb85e01d · outbound

This paper cites Common voice: A massively-multilingual speech corpus,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Common voice: A massively-multilingual speech corpus,

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.716163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation c18b0848-e607-4b93-8dfb-38506e54df6a · outbound

This paper cites V oxPopuli: A large-scale multilingual speech corpus for representation learning, semi-supervised learning and interpretation,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs V oxPopuli: A large-scale multilingual speech corpus for representation learning, semi-supervised learning and interpretation,

Reference 45

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.735106Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ec50311e-9717-4443-b3d6-a8e24d5cce5f · outbound

This paper cites Ted-lium 3: Twice as much data and corpus repartition for experiments on speaker adaptation,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Ted-lium 3: Twice as much data and corpus repartition for experiments on speaker adaptation,

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.697696Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation a3d1013e-9dd2-4e52-a9c8-ace355b79206 · outbound

This paper cites ”earnings- 22: A practical benchmark for accents in the wild.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs ”earnings- 22: A practical benchmark for accents in the wild

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.733023Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 8496c9cb-6b13-40a4-b382-7d1572013902 · outbound

This paper cites The ami meeting corpus: A pre-announcement,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs The ami meeting corpus: A pre-announcement,

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.739220Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation fb2b9101-cb2b-4880-9cb4-6c6957330305 · outbound

This paper cites FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs FLEURS: Few-shot Learning Evaluation of Universal Representations of Speech

Reference 49

Resolution: verified exact; local_arxiv, observed 2026-07-10T20:37:34.444265Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation b58aae29-800d-4348-878a-d7e8a6644d8e · outbound

This paper cites Robust speech recognition via large-scale weak super- vision,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Robust speech recognition via large-scale weak super- vision,

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.756790Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation 15296f6a-e947-4705-afe9-2de029452ce6 · outbound

This paper cites Conformer: Convolution- augmented Transformer for Speech Recognition,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs Conformer: Convolution- augmented Transformer for Speech Recognition,

Reference 51

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.750916Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Observation ed375f2a-eaa3-45b1-a3d6-2d3d93a3b60b · outbound

This paper cites LoRA: Low-Rank Adaptation of Large Language Models,.

Compress the Cache, Not the Speech Embedding: KV Compression for Efficient Speech LLMs LoRA: Low-Rank Adaptation of Large Language Models,

Reference 52

Resolution: verified fuzzy; raw_fallback, observed 2026-07-10T20:37:34.761787Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers