directed acyclic graph
PulseAugur coverage of directed acyclic graph — every cluster mentioning directed acyclic graph across labs, papers, and developer communities, ranked by signal.
- 2026-07-07 research_milestone Publication of a research paper introducing DAG, a new framework for 3D affordance learning using diffusion models. source
6 day(s) with sentiment data
-
New CAFE framework uses causal discovery to improve automated feature engineering
Researchers have developed CAFE, a novel framework for automated feature engineering (AFE) that integrates causal discovery with multi-agent reinforcement learning. Unlike traditional AFE methods that rely on statistica…
-
New system virtualizes LLM memory, reducing token counts by up to 86%
Researchers have developed a new system called Contextual Memory Virtualisation (CMV) to manage the growing state accumulated by large language models (LLMs) during extended reasoning tasks. CMV treats LLM understanding…
-
New framework enables autonomous code evolution for generative recommendation systems
Researchers have developed TGR-EvoDAG, a novel multi-agent framework designed for the autonomous evolution of code in Generative Recommendation (GR) systems. This Directed Acyclic Graph (DAG)-based approach allows for t…
-
New Metropolis-Hastings generalization improves DAG sampling for Bayesian networks
Researchers have developed a new method to improve the efficiency of approximating posterior distributions for directed acyclic graphs (DAGs) in Bayesian network structure learning. This advancement introduces novel Mar…
-
New graph-based method improves LLM response synthesis and reasoning accuracy
Researchers have developed Consensus Graphs (ConGrs), a novel data structure utilizing directed acyclic graphs to represent and synthesize information from multiple language model responses. This method improves factual…
-
Synthetic data generation needs lineage tracking, not just volume
The current approach to generating synthetic training data for AI models often overlooks the critical need for lineage tracking. While using a teacher model to create data is efficient, failing to record the specific te…
-
New framework for distributional regression using ReLU neural networks
This paper introduces a novel framework for distributional regression that accommodates arbitrary response variables. It models conditional event probabilities using measurable sets, extending the approach to general re…
-
LLM-guided framework optimizes DAG scheduling for heterogeneous systems
Researchers have developed MoSAIC, a novel framework for optimizing task scheduling in heterogeneous computing systems. This system utilizes large language models (LLMs) to learn effective scheduling heuristics by analy…
-
New DECO framework accelerates DAG learning by reducing search space
Researchers have developed a new framework called DECO (Directional Evidence-guided Configuration Optimization) to improve the efficiency of learning directed acyclic graphs (DAGs) from observational data. This non-para…
-
New method analyzes ODE identifiability with hidden confounders
Researchers have developed a new method for analyzing the identifiability of linear Ordinary Differential Equation (ODE) systems, particularly when hidden confounders are present. The paper addresses two scenarios: one …
-
New algorithm HOST improves Gaussian DAG learning efficiency
Researchers have developed HOST, a new algorithm for learning Gaussian Directed Acyclic Graphs (DAGs) that addresses the statistical-computational gap. Unlike existing methods that require computationally expensive subs…
-
New method ensures budget conservation in distributed AI agent delegation
Researchers have developed a new method for fault-tolerant budget conservation in distributed multi-agent delegation systems. This approach formalizes how budgets, represented by exclusive escrow credits, move through a…
-
New MatRAG framework enhances multi-hop QA with hierarchical RAG and MRL
Researchers have developed MatRAG, a novel hierarchical framework that integrates Retrieval-Augmented Generation (RAG) with Matryoshka Representation Learning (MRL) to improve multi-hop question answering systems. This …
-
New methods advance amortized Bayesian inference using neural networks · 2 sources tracked
Two new research papers submitted to arXiv introduce novel methods for amortized Bayesian inference. The first paper proposes using auxiliary supervision to mitigate representation gaps in neural network optimization fo…
-
New method boosts LLM causal reasoning with externalized CPDAG summaries
Researchers have developed a new method called Structured Thinking to improve the causal reasoning capabilities of large language models (LLMs). This two-turn pipeline first externalizes a summary of a Causal Probabilis…
-
New research advances causal discovery with LLMs and novel algorithms · 4 sources tracked
Recent research explores advanced techniques for causal discovery, a field focused on inferring cause-and-effect relationships from data. One paper investigates the assumptions embedded in simulated data used for superv…
-
New method refines AI agent trajectories, cutting costs and boosting accuracy
Researchers have developed a method called Dependency-Aware Trajectory Refinement (DATR) to optimize the fine-tuning of multi-turn AI agents. This technique involves representing agent trajectories as a Directed Acyclic…
-
New algorithm tackles causal discovery with latent confounders
Researchers have developed a new algorithm for causal discovery that can identify causal relationships among observed variables even when latent confounders are present. The algorithm works by reconstructing the precisi…
-
Agora system uses Git for collaborative AI research
Researchers have developed Agora, a system that uses Git as a shared memory for collective autonomous research. This system records research contributions as an append-only directed acyclic graph (DAG), allowing multipl…
-
New agent Intentest uses DAG to improve automated cybersecurity penetration testing
Researchers have developed Intentest, a novel agent designed for long-horizon automated penetration testing in cybersecurity. This agent addresses context forgetting and intent drift in LLM-based systems by externalizin…