Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:
ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Home/Authors/Ran Liu

Ran Liu

8 indexed papers

Recent (6 mo)
7
With code
0
Influential cites
0
Benchmarked
0

Publications per year

8
26

Top categories

ML×5AI×4NLP×2Info Retrieval×1Architecture×1Emerging Tech×1Software Eng.×1Crypto×1

Frequent co-authors

Haoran Liu3×
Hsin-Chiang Chen1×
Hao Wang1×
Gedi Zhou1×
Mingze Gao1×
Abhishek Kumar1×

Research Timeline

2026
ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts

The paper introduces SHADOWMERGE, a novel poisoning attack that successfully compromises graph-based agent memory by exploiting relation-channel conflicts, achieving a high attack success rate across multiple real-world benchmarks.

Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignment

The paper introduces AdvCL, a framework that repurposes adversarial perturbations as a geometric control signal to stabilize continual learning in large language models, significantly reducing forgetting and enhancing robustness.

Agents-K1: Towards Agent-native Knowledge Orchestration

This paper introduces Agents-K1, an end-to-end knowledge orchestration pipeline that converts raw documents into agent-native scientific knowledge graphs.

A Model-Driven Approach for Developing Families of Reinforcement Learning Environments

A model-driven approach is proposed for generating families of reinforcement learning training environments using a hybrid genetic algorithm and model transformations.

Leveraging ECRAM for Edge Continual Learning

The paper proposes CLASP, an end-to-end system with IMC acceleration for continual learning on edge platforms, addressing challenges of noisy computation and poor support for resource-efficient training.

Select or Rewrite: Oracle-Guided Fine-Tuning for Generative ASR Post-Editing from N-Best Hypotheses

The paper proposes oracle-guided fine-tuning to improve post-editing of ASR transcripts by using curated pools of corrections, regularizers, and hard negatives.

MemT-1.0: Towards Memory-Native Evolution

This paper introduces MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation.

Baseline and Incremental: User, Item, and Request Debiasing in Instagram Feed Ranking

Debiasing feed terms to improve engagement predictions

Highlighted terms show continued research focus across papers

Papers

cs.IRcs.LGEmpiricalRecentOct 20, 2026

Baseline and Incremental: User, Item, and Request Debiasing in Instagram Feed Ranking

Hsin-Chiang Chen, Hao Wang, Gedi Zhou, Mingze Gao +22 more

Debiasing feed terms to improve engagement predictions

View →
cs.CLcs.LGEmpiricalRecent
Oct 19, 2026

MemT-1.0: Towards Memory-Native Evolution

Zhiyu Li, Zeyu Zhang, Linyun Liu, Yiwei Bu +70 more

This paper introduces MemT-1.0, a new architecture of memory foundation models that internalize memory as part of model computation.

View →
cs.CLcs.AIEmpiricalRecentOct 10, 2026

Select or Rewrite: Oracle-Guided Fine-Tuning for Generative ASR Post-Editing from N-Best Hypotheses

Shiran Liu, Gaohui Xu, Sergio Burdisso, Thibault Bañeras-Roux +2 more

The paper proposes oracle-guided fine-tuning to improve post-editing of ASR transcripts by using curated pools of corrections, regularizers, and hard negatives.

View →
cs.ARcs.ETcs.LGEmpiricalRecentJul 22, 2026

Leveraging ECRAM for Edge Continual Learning

Nabila Tasnim, Haoran Liu, Qing Cao, Saugata Ghose

The paper proposes CLASP, an end-to-end system with IMC acceleration for continual learning on edge platforms, addressing challenges of noisy computation and poor support for resource-efficient traini…

View →
cs.SEcs.LGEmpiricalRecentJun 18, 2026

A Model-Driven Approach for Developing Families of Reinforcement Learning Environments

Xiaoran Liu, Istvan David

A model-driven approach is proposed for generating families of reinforcement learning training environments using a hybrid genetic algorithm and model transformations.

View →
cs.AIEmpiricalRecentJun 11, 2026

Agents-K1: Towards Agent-native Knowledge Orchestration

Zongsheng Cao, Bihao Zhan, Jinxin Shi, Jiong Wang +21 more

This paper introduces Agents-K1, an end-to-end knowledge orchestration pipeline that converts raw documents into agent-native scientific knowledge graphs.

View →
cs.LGcs.AIRecentJun 1, 2026

Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignment

Ran Liu, Min Yu, Mingqi Liu, Jianguo Jiang +6 more

The paper introduces AdvCL, a framework that repurposes adversarial perturbations as a geometric control signal to stabilize continual learning in large language models, significantly reducing forgett…

View →
cs.CRcs.AIMay 9, 2026

ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts

Yang Luo, Zifeng Kang, Tiantian Ji, Xinran Liu +3 more

The paper introduces SHADOWMERGE, a novel poisoning attack that successfully compromises graph-based agent memory by exploiting relation-channel conflicts, achieving a high attack success rate across…

View →