Autonomous LLM agents executing multi-step workflows over personal data often misalign with user privacy expectations and leak sensitive, out-of-context information. Although Contextual Integrity (CI) formalizes this problem, existing benchmarks and defenses rarely operationalize CI for realistic multi-step task sequen...
Robin Staab, Tom Wollschl\"ager, Vahab Mirrokni et al.· 0 citations
Evaluating the structural quality of candidate causal graphs when the true causal structure is unknown is an important problem in applying causal discovery results. This paper proposes CausalGrade, a causal graph scoring method based on the probabilistic judgment model Jev. The method jointly provides Jev with evidence...
Jiangming Zhong, Jiawei Zhu, Guanting Wu et al.· 0 citations
AlphaFold 3-like models (e.g., Boltz, Protenix, etc.) have pushed the frontier of protein binder design. This architecture is used both for the generation of binder candidates (e.g., by hallucination or converting these models into backbone or full atom generative models) as well as for the screening of these candidate...
Lorenz Vaitl, Ivan Tanasijevic, Arian Rokkum Jamasb et al.· 0 citations
Recent work has shown that gradient updates in federated learning (FL) can unintentionally reveal sensitive information about a client's local data. This risk becomes significantly greater when a malicious server manipulates the global model to provoke information-rich updates from clients. In this paper, we adopt a de...
Agentic workflows are increasingly shifting towards long-horizon tasks, such as open-ended scientific work, where models benefit from long context windows. However, at ultra-long context (above 1M tokens), full attention is prohibitively expensive. A promising approach is sparse attention, in which a dedicated indexer...
Ales Kubicek, Andrea Caciolai, Liangzu Peng et al.· 0 citations
In real-world deployments, large language model (LLM) systems are increasingly evolving from single-model invocation toward heterogeneous deployments in which different models are dynamically invoked for inference. When the same context is reused across models, however, each newly selected target model typically has to...
CWAM (Causal WAM), which incorporates the sparsity constraint into a pretrained WAM by adding a sparse routing path to its action head, with the world modeling objective remaining unchanged, is proposed.
Jun-Tao Cheng, Jin-Zhou Tang, Fan Feng et al.· 0 citations
AI answers require more than apparent correctness: reviewers need to inspect their recorded basis. We introduce the Knowledge Layer Model (KLM), an open, implementation-neutral specification for inference records with versioned epistemic assessments, evidence relations, and explicit unknowns. SEDIM, the knowledge and e...
An audit of widely deployed LLMs from four providers on smoking cessation, a domain with codified clini- cal guidance and substantial public-health stakes, by putting the same four cessation questions to twelve user personas that differ in life stage, self-described expertise, and smoking history, for a total of 2,880...
Artin Yousefi, Aizierjiang Aiersilan, L. Abroms et al.· Proceedings of the AAAI/ACM...· 0 citations
TACTIC (Trade-aware Adaptive Consolidation and Theory-of-mind for International Commerce), an agent that iteratively refines its negotiation strategy across episodes-without updating base LLM weights-by leveraging prior experience, opponent behavior, and past settlement outcomes is proposed.
Linear attention scales linearly with sequence length, but its finite-state memory can limit long-context retrieval. In gated linear recurrences, memory retention depends on state decay, which is shaped jointly by gate parameters A_log and dt_bias and by input activations. In controlled Gated DeltaNet (GDN) experiments...
Zhang-Chi Feng, Harry Jiang, Yangyifei Luo et al.· 0 citations
