Skip to content

Research

Page 65 of 88

Generating Proof-of-Vulnerability Tests to Help Enhance the Security of Complex Software

research note

Generating Proof-of-Vulnerability Tests to Help Enhance the Security of Complex Software

·7 min read·Shravya Kanchi, Xiaoyan Zang, Ying Zhang et al.

This paper addresses the challenge of automatically generating proof-of-vulnerability (PoV) tests to demonstrate real exploitability of vulnerabilities in third-party libraries (Libs) as they affec…

researchproof-of-vulnerabilitysoftware-supply-chain-securityai-assisted-test-generationcall-path-analysis

Read note → Source paper ↗

Label-Efficient School Detection from Aerial Imagery via Weakly Supervised Pretraining and Fine-Tuning

research note

Label-Efficient School Detection from Aerial Imagery via Weakly Supervised Pretraining and Fine-Tuning

·9 min read·Zakarya Elmimouni, Fares Fourati, Mohamed-Slim Alouini

This work addresses the challenge of detecting school infrastructure from aerial imagery in contexts with scarce manual annotations, a critical problem for global education and connectivity initiat…

researchweakly-supervised-learningremote-sensingobject-detectionaerial-imagery

Read note → Source paper ↗

PHALAR: Phasors for Learned Musical Audio Representations

research note

PHALAR: Phasors for Learned Musical Audio Representations

·8 min read·Davide Marincione, Michele Mancusi, Giorgio Strano et al.

This paper addresses the problem of stem retrieval in musical audio, where given a partial mix, the goal is to identify the missing complementary stems that coherently fit it both harmonically and …

researchaudio-representation-learningcontrastive-learningmusic-information-retrievalcomplex-valued-neural-networks

Read note → Source paper ↗

Precomputed Lens Transport Maps

research note

Precomputed Lens Transport Maps

·10 min read·Yang Chen, Xiaochun Tong, Afet Abzar et al.

This paper addresses a practical gap in camera/lens simulation: existing fast approximations either model only geometry (pinhole, thin lens, ABCD, polynomial optics) or only the smooth refraction c…

researchcomputer-graphicsneural-renderingray-tracinglens-simulation

Read note → Source paper ↗

Robustness and Transferability of Pix2Geomodel for Bidirectional Facies Property Translation in a Complex Reservoir

research note

Robustness and Transferability of Pix2Geomodel for Bidirectional Facies Property Translation in a Complex Reservoir

·8 min read·Abdulrahman Al-Fakih, Nabil Sariah, Ardiansyah Koeshidayatullah et al.

This paper investigates the robustness and transferability of Pix2Geomodel, a conditional GAN-based framework for bidirectional translation between geological facies and petrophysical properties (p…

researchreservoir-geomodelingconditional-ganimage-to-image-translationpix2pix

Read note → Source paper ↗

Stayin' Aligned Over Time: Towards Longitudinal Human-LLM Alignment via Contextual Reflection and Privacy-Preserving Behavioral Data

research note

Stayin' Aligned Over Time: Towards Longitudinal Human-LLM Alignment via Contextual Reflection and Privacy-Preserving Behavioral Data

·12 min read·Simret Araya Gebreegziabher, Allison E Sproul, Yinuo Yang et al.

This paper attacks a foundational assumption baked into RLHF and most preference-learning pipelines: that a user's evaluation of an LLM response, captured immediately after generation, is a stable …

researchhuman-ai-alignmentlongitudinal-evaluationpreference-learningrlhf

Read note → Source paper ↗

Towards Open World Sound Event Detection

research note

Towards Open World Sound Event Detection

·9 min read·P. H. Hai, L. T. Minh, L. H. Son

This paper argues that standard sound event detection is too closed-world for real deployments: at inference, audio scenes contain novel, unlabeled, and often overlapping events that current detect…

researchsound-event-detectionopen-world-learningincremental-learningtransformers

Read note → Source paper ↗

cotomi Act: Learning to Automate Work by Watching You

research note

cotomi Act: Learning to Automate Work by Watching You

·14 min read·Masafumi Oyamada, Kunihiro Takeoka, Kosuke Akimoto et al.

cotomi Act tackles two distinct but complementary deficiencies in current computer-using agents (CUAs): unreliable multi-step web task execution, and complete ignorance of the tacit organizational …

researchbrowser-automationcomputer-using-agentllm-agentorganizational-knowledge

Read note → Source paper ↗

Articles are CC BY 4.0 — feel free to quote with attribution