{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","short_pith_number":"pith:R6GACQN7","canonical_record":{"source":{"id":"2607.26633","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","cross_cats_sorted":[],"title_canon_sha256":"0a93a664eca161a8398332b68f6510332976170ad3ed037df33e364b10c27c11","abstract_canon_sha256":"54ca0fe4ce2caa61a65060b4168ab158bd5174b7e33e762d87bd51d7b1fab178"},"schema_version":"1.0"},"canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","source":{"kind":"arxiv","id":"2607.26633","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26633","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26633v1","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26633","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_12","alias_value":"R6GACQN7Z5OB","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_16","alias_value":"R6GACQN7Z5OB3VUC","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_8","alias_value":"R6GACQN7","created_at":"2026-07-30T01:21:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","target":"record","payload":{"canonical_record":{"source":{"id":"2607.26633","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","cross_cats_sorted":[],"title_canon_sha256":"0a93a664eca161a8398332b68f6510332976170ad3ed037df33e364b10c27c11","abstract_canon_sha256":"54ca0fe4ce2caa61a65060b4168ab158bd5174b7e33e762d87bd51d7b1fab178"},"schema_version":"1.0"},"canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","last_reissued_at":"2026-07-30T01:21:38.317044Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:21:38.317044Z"},"source_kind":"arxiv","source_id":"2607.26633","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:21:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"idu3VE9F7xPL+VdoIMI+PYy2NqQiXm1+XokPPdCxZ6JhmA43GUgdJZBQwMn2kBNxlL9wFQMRVYHlwXUoNccvAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:36:24.267372Z"},"content_sha256":"702802e59e0c3a469584249795ecfdc433671b800878104f117e2f7a10704e65","schema_version":"1.0","event_id":"sha256:702802e59e0c3a469584249795ecfdc433671b800878104f117e2f7a10704e65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"NELSSA: A GPU-PNM Heterogeneous System for Mixed-Length LLM Serving via Length-based Request Placement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Donguk Moon, Euiseok Kim, Guseul Heo, Hoshik Kim, Hyeongseok Gwak, Jongse Park, Junseok Lee, Kangkyu Park, Kwangsik Shin, Myunghyun Rhee, Seungyong Lee, Sookyung Choi, Youngpyo Joo, Yunseo Chun","submitted_at":"2026-07-29T08:58:04Z","abstract_excerpt":"Modern LLMs and their agentic applications are broadening the range of serving workloads, spanning context lengths from a few hundred tokens to hundreds of thousands. As these requests frequently interleave within the same serving window, LLM serving systems must handle highly heterogeneous mixed-length workloads. Such mixed-length workloads expose fundamental inefficiencies in GPU-centric serving architectures, whose throughput depends on large, memory-constrained batches. In this paper, we present NELSSA, an LLM serving system that integrates GPUs with real-world Processing-near-Memory (PNM)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26633","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.26633/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:21:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TyhpehTSrOvww3tqzLEsxN2ZNUl+r/AEknsr9A43vkWc/y58sFxU/dNwXLuuQI+i1C5AcYfl0B7JvJa2A1MmAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:36:24.267908Z"},"content_sha256":"d5e157c94de0588598ffa28d064e7e7ec0c27cc88d417f6ade85fd4f10af1aef","schema_version":"1.0","event_id":"sha256:d5e157c94de0588598ffa28d064e7e7ec0c27cc88d417f6ade85fd4f10af1aef"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.57967/hf/3372) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Leonid Pekelis, Michael Feil, Forrest Moret, Mark Huang, and Tiffany Peng. 2024. Llama 3 Gradient: A series of long context models. https://doi.org/10.57967/hf/ 3372","arxiv_id":"2607.26633","detector":"doi_compliance","evidence":{"ref_index":38,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"10.57967/hf/","reconstructed_doi":"10.57967/hf/3372"},"severity":"advisory","ref_index":38,"audited_at":"2026-08-01T12:08:46.539718Z","event_type":"pith.integrity.v1","detected_doi":"10.57967/hf/3372","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"5b41313b529ac189f63c4d25430e21a15c97faca7c1e8e243a09e85cd4df2507","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15764,"payload_sha256":"64b5b919f81d4a0f96156b7ed5c4324b1bec3d6f767ecde311e6f109499b31b4","signature_b64":"zPhrRweIBkk35OtNfcqoc2tPwoNViDfTXR5lgK5/2VGU7xaxqEqqqhMrNVDa3f47zmlU7Axe2r7nBuA0O3xHCg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T12:13:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CDZ2VyjRPys2jJcmUUP9RaI77e0Uzqlchqwe/KUmOUghpQTKE/byx/A9nroWf7Ui4VSnLi6rWF55jFgmZ9LgBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T09:36:24.271272Z"},"content_sha256":"172acc98a46bec08e26d94bd5690af3f924d2c35f7e10f7d05e717d9977d349a","schema_version":"1.0","event_id":"sha256:172acc98a46bec08e26d94bd5690af3f924d2c35f7e10f7d05e717d9977d349a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/bundle.json","state_url":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T09:36:24Z","links":{"resolver":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE","bundle":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/bundle.json","state":"https://pith.science/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/R6GACQN7Z5OB3VUCJ3UPDSCPXE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:R6GACQN7Z5OB3VUCJ3UPDSCPXE","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"54ca0fe4ce2caa61a65060b4168ab158bd5174b7e33e762d87bd51d7b1fab178","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","title_canon_sha256":"0a93a664eca161a8398332b68f6510332976170ad3ed037df33e364b10c27c11"},"schema_version":"1.0","source":{"id":"2607.26633","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26633","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26633v1","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26633","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_12","alias_value":"R6GACQN7Z5OB","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_16","alias_value":"R6GACQN7Z5OB3VUC","created_at":"2026-07-30T01:21:38Z"},{"alias_kind":"pith_short_8","alias_value":"R6GACQN7","created_at":"2026-07-30T01:21:38Z"}],"graph_snapshots":[{"event_id":"sha256:d5e157c94de0588598ffa28d064e7e7ec0c27cc88d417f6ade85fd4f10af1aef","target":"graph","created_at":"2026-07-30T01:21:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.26633/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Modern LLMs and their agentic applications are broadening the range of serving workloads, spanning context lengths from a few hundred tokens to hundreds of thousands. As these requests frequently interleave within the same serving window, LLM serving systems must handle highly heterogeneous mixed-length workloads. Such mixed-length workloads expose fundamental inefficiencies in GPU-centric serving architectures, whose throughput depends on large, memory-constrained batches. In this paper, we present NELSSA, an LLM serving system that integrates GPUs with real-world Processing-near-Memory (PNM)","authors_text":"Donguk Moon, Euiseok Kim, Guseul Heo, Hoshik Kim, Hyeongseok Gwak, Jongse Park, Junseok Lee, Kangkyu Park, Kwangsik Shin, Myunghyun Rhee, Seungyong Lee, Sookyung Choi, Youngpyo Joo, Yunseo Chun","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","title":"NELSSA: A GPU-PNM Heterogeneous System for Mixed-Length LLM Serving via Length-based Request Placement"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26633","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:702802e59e0c3a469584249795ecfdc433671b800878104f117e2f7a10704e65","target":"record","created_at":"2026-07-30T01:21:38Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"54ca0fe4ce2caa61a65060b4168ab158bd5174b7e33e762d87bd51d7b1fab178","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-07-29T08:58:04Z","title_canon_sha256":"0a93a664eca161a8398332b68f6510332976170ad3ed037df33e364b10c27c11"},"schema_version":"1.0","source":{"id":"2607.26633","kind":"arxiv","version":1}},"canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8f8c0141bfcf5c1dd6824ee8f1c84fb931c50e0e6e39de51b717831ad46327b9","first_computed_at":"2026-07-30T01:21:38.317044Z","kind":"pith_receipt","last_reissued_at":"2026-07-30T01:21:38.317044Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.26633","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:702802e59e0c3a469584249795ecfdc433671b800878104f117e2f7a10704e65","sha256:d5e157c94de0588598ffa28d064e7e7ec0c27cc88d417f6ade85fd4f10af1aef","sha256:172acc98a46bec08e26d94bd5690af3f924d2c35f7e10f7d05e717d9977d349a"],"state_sha256":"ce4992724e4285a2e2962031269ef5282c467152c8c3f606d7b1a241ab883617"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g+85o1cUdgtWKF3ap4E4K1aZnmlH/9cyY+XYqQwt0EgczaI49qtTQVMD+TqJ9YOu2ReaQZ0ATTVBpWK4NixXBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T09:36:24.273326Z","bundle_sha256":"c0622565019dfb8b7ac038ad5b738970b038914bdddd2cfe748c80052ef27bf9"}}