arXiv cs.CL
TIER_1
English(EN)
·
John Sweeney
·
2026-07-01 04:00
arXiv:2606.30788v1 Announce Type: cross Abstract: Language models are often adapted in stages: a public skill phase, a private memory phase, and a later safety phase that learns to refuse outputs tied to the remembered entities. Revoking the memory after the safety phase is not t…
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
8/ Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining (OEA)
Paper: https://t.co/dw33plIoxW
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
6/ When RL Meets Adaptive Speculative Training: A Unified Training-Serving System (Aurora)
Paper: https://t.co/fvLuHrqDbX
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
7/ Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking
Paper: https://t.co/LgGqu8vl97
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
5/ V1: Unifying Generation and Self-Verification for Parallel Reasoners
Paper: https://t.co/X1zUsS7gY8
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
2/ ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System
Paper: https://t.co/7I1Yf5s8B8
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
3/ Learning to Discover at Test Time (TTT-Discover)
Paper: https://t.co/pKeadv4DHl
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
4/ Escaping the Verifier: Learning to Reason via Demonstrations (RARO)
Paper: https://t.co/gQZCEav8Nb
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-07-01 04:34
1/ DSGym: A Holistic Framework for Evaluating and Training Data Science Agents
Paper: https://t.co/jV4uMB1g48
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
8/ Opportunistic Expert Activation: Batch-Aware Expert Routing for Faster Decode Without Retraining (OEA)
https://t.co/dw33plIoxW
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
5/ V1: Unifying Generation and Self-Verification for Parallel Reasoners
https://t.co/X1zUsS7gY8
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
6/ When RL Meets Adaptive Speculative Training: A Unified Training-Serving System (Aurora)
https://t.co/fvLuHrqDbX
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
7/ Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking
https://t.co/LgGqu8vl97
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
2/ ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System
https://t.co/7I1Yf5s8B8
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
3/ Learning to Discover at Test Time (TTT-Discover)
https://t.co/pKeadv4DHl
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
4/ Escaping the Verifier: Learning to Reason via Demonstrations (RARO)
https://t.co/gQZCEav8Nb
X — Together (inference / OSS)
TIER_1
English(EN)
·
togethercompute
·
2026-06-30 19:54
1/ DSGym: A Holistic Framework for Evaluating and Training Data Science Agents
https://t.co/jV4uMB1g48