Skip to content

Research separate from news

What could become important next

A preprint is a signal, not a finished product or an independently confirmed result. We therefore track research papers separately and show the actual state of supporting evidence for each one.

1381 published research events

Latest work

A significant claim from a single source is published only after further confirmation.

Research only one source so far

Study introduces perturber, a tool for testing changes to input text

The authors introduce perturber, a service and Python library with 68 text transformations in 11 groups. Each call applies one transformation. The study evaluates meaning preservation, reproducibility and the cost of using the tool; it does not test model robustness.

arXiv cs.CL (Computation and Language / NLP) Original source ↗
Research only one source so far

Study introduces HMP-GAE method for protecting federated LLM training

The HMP-GAE method identifies and suppresses semantically inconsistent updates during federated LLM training. In tests, the authors report an increase in accuracy of 1.21 %, a reduction in semantic entropy of 44.08 % and a reduction in perplexity of 13.34 % compared with existing defenses.

arXiv cs.LG (Machine Learning) Original source ↗
Research only one source so far

Study evaluates PRISM model for clinical prognosis using different types of health data

The authors present PRISM, a model that combines measurements, ECG, medical history and population data. Using only inputs within the benchmark’s time window, it achieved an AUROC of 0.8895 compared with 0.8937 for UniPACT. Higher scores rely on pre-arrival history that was not available to the models used for comparison.

arXiv cs.AI (Artificial Intelligence) Original source ↗
Research only one source so far

Study modifies how ResonatorLM reads memory for longer sequences

The authors condition the memory-reading stage on the current input, with a 0.2 % increase in the parameter count. For models with 6 million parameters trained on sequences of length 256 and tested on sequences of length 1024, perplexity fell from 12.81 to 3.76, according to the study. Information recall did not improve.

arXiv cs.LG (Machine Learning) Original source ↗
Research only one source so far

The CLUE framework automates the creation and annotation of event types in texts

The study introduces the CLUE framework for creating, naming and annotating event types without manually designing an ontology. It combines representations of verbs and event meanings with contrastive learning and clustering. The authors report that it is effective in experiments and that the code is available.

arXiv cs.CL (Computation and Language / NLP) Original source ↗
Research only one source so far

Study introduces SELFCOMPACT to manage context reduction for AI agents

The study introduces SELFCOMPACT: the model shortens its context using a tool and rules for its use. Across six task sets and seven models, the authors report an improvement of up to 18.1 points in mathematics compared with a version without summarization, with per-question costs reduced by 30–70 %.

Apple Machine Learning Research Original source ↗
Research only one source so far

Study shows the limitations of both LLMs and humans in assessing machine translation quality

The study compared translation assessments by LLMs and humans using a dataset with 70 language pairs and data from WMT23 and WMT25. According to the authors, agreement is unreliable under most conditions; LLMs struggle particularly with minor errors, language varieties and identifying erroneous text spans.

Apple Machine Learning Research Original source ↗
Research only one source so far

Study introduces Synthesis Through Simulation for generating enterprise data without database schemas

Synthesis Through Simulation is a method that generates synthetic enterprise data using an agent that performs operations through APIs in simulated environments. The authors report an average marginal distribution fidelity of 0.88 and satisfaction of 100 % of constraints across ten environments without access to database schemas.

arXiv cs.AI (Artificial Intelligence) Original source ↗
Research only one source so far

In the StoreBench study, AI agents lagged behind a rule-based strategy in store management

The authors introduce StoreBench for evaluating agents in simulated store management. Of the seven models, DeepSeek-V4-Pro achieved the highest success rate: 49 % compared with 97 % for a rule-based strategy. They publish sample tasks and evaluation tools, while the full environment remains unavailable to the public.

arXiv cs.AI (Artificial Intelligence) Original source ↗