A benchmark of 400 multiple-choice questions that combines widely known inside-knowledge inside-knowledge questions, which should be answered directly, with outside-knowledge questions about facts from news published after the models'training cut-offs is introduced, which should trigger search or, without tools, absten...
PACE (Proxy Alignment and Curvature Expansion), a pre-optimization diagnostic that uses repeated human ratings to predict the human-gain curve induced by a fixed proxy, characterizes gentle optimization along the positive tilt branch of a fixed item pool.
Quantization of Large Language Models (LLMs) has be- coming increasingly important for reducing the computa- tional costs and resource requirements for practical deploy- ment.Growinginterestinthisareahasmotivatedthecommu- nity to move beyond 8-bit quantization towards 4-bit preci- sion, leading to the development of cu...
Jing Huang, Yu Cheng, Ziwei Yu et al.· 0 citations
Manifold learning methods like t-SNE and UMAP preserve local neighborhoods, but they do not preserve the ranking of the neighbors inside them. Two embeddings can return the same neighbor set while disagreeing on which point is closest. This loss of order degrades downstream applications that value the order in the embe...
Henry Han, Luyu Niu, Diane Li et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Time-series foundation models (TSFMs) provide zero-shot forecasts for unseen series, but their accuracy varies across datasets. Combining several models can reduce this variability, yet zero-shot deployment provides no target labels for selecting or weighting them. We find that a model pool is useful not only as a fina...
Junhyeok Kang, Sangjun Han, Jun Seo et al.· 0 citations
In text-to-SQL and AI-agent analytics, successful SQL execution does not establish semantic correctness under changing business definitions. We study revalidation of fixed queries after changes to refund attribution, tax treatment, eligible populations and exchange-rate dates. An executable verification profile binds t...
Gen-Liang Zhu, Xiao-Bo Chen, Zi-Yu Wang et al.· 0 citations
CIRCE is introduced, which extracts a problem's structure by using a cheap LLM to construct a Constraint Intermediate Representation that deterministically compiles into the constraint graph.
Chen-Hao Hong, Jia-Kai Xu, Kaushiki Singh et al.· 0 citations
These results separate multi-hop candidate discovery, record-based discrimination and review prioritization within an explicit synthetic evaluation design for payroll-fraud screening from record-aware rules.
Xiao-Bo Chen, Gen-Liang Zhu, Zi-Yuan Wang et al.· 0 citations
The results demonstrate the effectiveness of complementary optimization strategies and the interpretability of optimal vocabulary-based discrimination in the MRBSSP.
Marko Djukanović, Staša Vujičić Stanković, Žiga Zebec et al.· 0 citations
The BABA-Bench, a benchmark of 47 interactive worlds, built from a de-identified, real patient advocacy case that expert advocates reviewed, is introduced, a benchmark of 47 interactive worlds for LLM agents.
S. Vasudevan, Videet Mehta, Connor Sweeney· 0 citations
The work demonstrates LanG's practical scalability, processing 3.33 million records in 6.4 minutes, over 30 million hourly, and its safeguards resisted prompt injection, label poisoning and ledger tampering, so small local models can carry most SOC detection affordably and safely.
Anes Abdennebi, Nadjia Kara, Laaziz Lahlou et al.· 0 citations
It is found that LLM team with a dedicated external conflict resolution protocol results in fewest breaches, and is also found that LLM team with a dedicated external conflict resolution protocol results in fewest breaches.
Khawaja Abaid Ullah, A. Acharya, M. J. Khojasteh· 0 citations
