llm-kb
← Back to research
Paper

SCENT: Language-guided framework bridges vision and olfaction using VLMs

Researchers propose SCENT, a multimodal framework that uses language guidance from Vision-Language Models (VLMs) to align visual scenes with olfactory signals. This addresses the challenge that many olfactory cues arise from contextual environmental factors not directly visible in pixels.

0 engagement·1 source·Tue, Jul 7, 2026, 03:31 PM
The paper 'What Images Cannot Say: Language-Guided Olfactory Representation Learning' introduces SCENT, which leverages VLMs to generate scene descriptors capturing objects, environmental context, and plausible ambient smells. This approach bridges the gap between vision and olfaction by using language as a semantic bridge, enabling better alignment of smell signals with images from electronic-nose measurements.

Entities

SCENT(concept)Vision-Language Models(concept)arXiv(tool)

Related

PaperTue, Jul 7, 2026, 05:27 PM

VAORA reward design addresses VLM failures in interactive physical reasoning

A new paper on arXiv introduces VAORA (Visual Action Outcome Reasoning Alignment), a reward design that targets two key failure modes in vision-language models: hallucinated chain-of-thought reasoning and misalignment between reasoning and actions. VAORA uses a Visual Alignment Reward to anchor reasoning to visual context, aiming to improve generalization in unseen interactive physical reasoning tasks.

0 engagement·1 source·arxiv
arXiv
Tool ReleaseMon, Jul 6, 2026, 09:30 PM

LocalEyes gives blind LLMs vision via local Ollama models

LocalEyes is a new tool that enables text-only LLMs like DeepSeek, CodeLlama, and Qwen-Coder to process images locally using an Ollama vision model. It supports screen capture, clipboard reading, and image file analysis without cloud uploads or API keys, offering a private, fast, and free solution for developers using Claude Code.

26 engagement·1 source·github
ProductSun, Jul 12, 2026, 04:39 PM

NeatContext: lightweight desktop app to give LLMs domain knowledge for oncall incident handling

NeatContext is a desktop application that lets LLMs access domain knowledge to handle oncall incidents more accurately. It solves the problem of SRE agents lacking domain-specific context, enabling better incident response without heavy infrastructure.

4 engagement·1 source·hackernews
PaperTue, Jul 7, 2026, 05:43 PM

Researchers identify modality interference as root cause of full-duplex SLM degradation

A new paper on arXiv (July 7, 2026) presents a fine-grained analysis of optimization dynamics in full-duplex Spoken Language Models (SLMs), identifying severe modality interference as the root cause of knowledge degradation and compromised semantic integrity. The work aims to enable more natural and intelligent full-duplex SLMs.

0 engagement·1 source·arxiv
arXiv
PaperSun, Jul 12, 2026, 08:50 AM

Researcher seeks arXiv endorsement for multi-agent citation verification framework

A researcher is seeking an arXiv endorsement for a paper proposing a four-agent framework built on CrewAI that addresses hallucinated citations in LLM-generated literature reviews. The framework includes an Academic Retriever, Critical Reviewer, Technical Writer, and Editor/Verifier implementing claim-level citation verification.

3 engagement·1 source·reddit