Sage
PulseAugur coverage of Sage — every cluster mentioning Sage across labs, papers, and developer communities, ranked by signal.
- 2026-07-21 research_milestone Researchers published a paper introducing SAGE, a new planning paradigm for latent world models that improves long-horizon planning. source
- 2026-05-25 research_milestone Publication of a research paper detailing the SAGE system for language-conditioned drone exploration and mapping. source
- 2026-05-18 research_milestone Publication of a new framework, SAGE, designed to improve spatial reasoning in Vision-Language Models. source
- 2026-05-08 research_milestone Publication of a new research paper introducing the SAGE method for multi-distribution learning. source
12 day(s) with sentiment data
-
New framework enhances LLM empathy using verifiable emotion feedback
Researchers have developed a dual-loop self-evolution framework to improve the empathetic capabilities of large language models in multi-turn dialogues. This framework uses verifiable emotion feedback to train the dialo…
-
Neurosymbolic AI Discovers Algebraic Graph Constructions
Researchers have developed a neurosymbolic agent capable of discovering algebraic descriptions for graphs, even when provided only with raw data like adjacency matrices. This agent combines a large language model with t…
-
SAGE system optimizes RAG retrieval for latency and cost · 2 sources tracked
Researchers have developed SAGE, a new adaptive retrieval policy for production Retrieval-Augmented Generation (RAG) systems. SAGE dynamically adjusts the number of passages retrieved per query based on estimated query …
-
New SAGE framework offers semantic explanations for AI in pathology
Researchers have developed SAGE, a new post-hoc framework designed to provide semantic, language-grounded explanations for attention-based multiple instance learning (ABMIL) models used in computational pathology. Unlik…
-
New SAGE dataset targets AI bias in South Asian medical imaging
Researchers have introduced SAGE, a new dataset of 1,300 expert-annotated GI endoscopy images from the South Asian region. This dataset aims to address the underrepresentation of diverse geographic populations in existi…
-
New attack method bypasses AI jailbreak defenses by combining code and search techniques
Researchers have demonstrated that a combination of a code-completion encoding and a best-of-N search can effectively bypass self-check jailbreak defenses in AI models, achieving success rates as high as 67% on some tar…
-
New SAGE architecture prioritizes AI safety over utility
A new research paper introduces SAGE, a safety-first architecture designed to control high-impact generative AI throughout its lifecycle. SAGE prioritizes safety over utility and commercial objectives, employing methods…
-
New LLM framework SAGE optimizes ad auto-bidding with parameter efficiency
Researchers have developed SAGE, a new framework for LLM-based auto-bidding in advertising auctions. SAGE utilizes a parameter-efficient multi-modal alignment approach to adapt LLMs for constrained bidding environments.…
-
Stable Diffusion user shares working Sage and Flash Attention 2 setup for Windows
A user on Reddit has successfully integrated Sage and Flash Attention 2 into their Windows 11 ComfyUI installation for Stable Diffusion. They shared a working configuration including specific software versions and hardw…
-
New tokenizer method improves cross-lingual homograph handling in LLMs
Researchers have identified a limitation in multilingual language models where shared subword vocabularies can lead to identical surface forms being treated too uniformly across languages, even when their meanings diffe…
-
New research improves latent world models for robotics planning · 2 sources tracked
Two new research papers introduce novel approaches to enhance latent world models for improved planning in robotics and control tasks. The first paper, DWM, proposes a method to separate action-driven transitions from a…
-
New SAGE method improves multi-distribution learning by considering flatness and gradient alignment
Researchers have introduced SAGE (Spectral-Aware Gradient-Aligned Exploration), a novel method for multi-distribution learning that addresses limitations in existing approaches. Unlike methods that focus solely on flatn…
-
Finance professionals spend 13 hours weekly verifying AI outputs, Sage study finds
New research from Sage indicates that finance professionals dedicate approximately 13 hours each week to verifying AI-generated outputs. This finding underscores the growing importance of trust, transparency, and govern…
-
New SPL language unifies LLM orchestration and symbolic computation
Researchers have introduced SPL (Structured Prompt Language), a novel declarative language designed to unify deterministic and probabilistic computation within a single specification. Unlike existing frameworks that sep…
-
SAGE method enhances visual place recognition with spatial-visual graph exploration
Researchers have developed SAGE (Spatial-visual Adaptive Graph Exploration), a novel training pipeline designed to improve visual place recognition. This method enhances the discrimination of local visual features by dy…
-
New SAGE framework unifies action and gaze recognition and anticipation
Researchers have introduced SAGE, a novel framework designed to simultaneously recognize and anticipate human actions and gaze patterns. Unlike previous models that treated these elements separately, SAGE integrates the…
-
New research tackles Text-to-SQL efficiency, correctness, and vulnerabilities
Researchers are developing new methods to improve the efficiency and reliability of Text-to-SQL systems. One approach, SQuaD-SQL, uses LLM-guided knowledge distillation to enable smaller language models to achieve high …
-
AI agents tackle complex math problems, setting new research benchmarks · 8 sources tracked
Researchers are developing advanced AI agents capable of tackling complex mathematical problems, pushing the boundaries of automated reasoning. Systems like ProofCouncil and OpenProver are demonstrating significant capa…
-
New protocol audits AI agents for accuracy and cost savings
Researchers have developed a new protocol for evaluating agentic Video Question Answering (VideoQA) systems, focusing on both accuracy and cost. This method pairs two systems to jointly assess differences in correctness…
-
AI talks on multi-agent systems and LLM guardrails scheduled for July 16
The North East Data Scientists group is hosting two talks on July 16th, focusing on advancements in AI. The first presentation will cover multi-agent AI workflows, while the second will explore open-source guardrails fo…