Natural language understanding (NLU) is integral to various social media applications. However, existing NLU models rely heavily on context for semantic learning, resulting in compromised performance when faced with short and noisy social media content. To address this issue, we leverage in-context learning (ICL), wherein language models learn to make inferences by conditioning on a handful of demonstrations to enrich the context and propose a novel hashtag-driven in-context learning (HICL) framework. Concretely, we pre-train a model #Encoder, which employs #hashtags (user-annotated topic labels) to drive BERT-based pre-training through contrastive learning. Our objective here is to enable #Encoder to gain the ability to incorporate topic-related semantic information, which allows it to retrieve topic-related posts to enrich contexts and enhance social media NLU with noisy contexts. To further integrate the retrieved context with the source text, we employ a gradient-based method to identify trigger terms useful in fusing information from both sources. For empirical studies, we collected 45M tweets to set up an in-context NLU benchmark, and the experimental results on seven downstream tasks show that HICL substantially advances the previous state-of-the-art results. Furthermore, we conducted extensive analyzes and found that: (1) combining source input with a top-retrieved post from #Encoder is more effective than using semantically similar posts; (2) trigger words can largely benefit in merging context from the source and retrieved posts.
[https://arxiv.org/abs/2308.09985v1]
Today's relaxing FLAC hacking session: Adding #metadata
writing to #SerenityOS's #FLAC #encoder!
#metadata #serenityos #flac #encoder
Long-term Forecasting with TiDE: Time-series Dense Encoder
Abhimanyu Das, Weihao Kong, Andrew Leach, Shaan K Mathur, Rajat Sen, Rose Yu
Action editor: Alessandro Sperduti.
#forecasting #encoder #transformer
Long-term Forecasting with TiDE: Time-series Dense Encoder
#forecasting #encoder #transformer
Relating graph auto-encoders to linear models
Sequential Query Encoding for Complex Query Answering on Knowledge Graphs
Jiaxin Bai, Tianshi Zheng, Yangqiu Song
Action editor: Alessandro Sordoni.
Get that AV1 encoding on the cheap!
Intel's Arc A750 beats the RTX 3060 - and is $200 at Newegg https://www.rockpapershotgun.com/intels-arc-a750-beats-the-rtx-3060-and-is-200-at-newegg
#Intel #Arc #A750 #RTX3060 #Deal #Newegg #AV1 #Encoder #GPU #TechNews
#intel #arc #a750 #rtx3060 #deal #newegg #av1 #encoder #gpu #technews
CAE v2: Context Autoencoder with CLIP Latent Alignment
Reducing Predictive Feature Suppression in Resource-Constrained Contrastive Image-Caption Retrieval
Maurits Bleeker, Andrew Yates, Maarten de Rijke
High Fidelity Neural Audio Compression
Kino Wheels Gives You a Hand Learning Camera Operation https://hackaday.com/2023/03/28/kino-wheels-gives-you-a-hand-learning-camera-operation/ #ArduinoMega2560 #MiscHacks #handwheel #simulator #bearing #encoder #Tilt #din #pan
#arduinomega2560 #MiscHacks #handwheel #simulator #bearing #encoder #tilt #din #pan
Kino Wheels Gives You a Hand Learning Camera Operation - Have you ever watched a movie or a video and really noticed the quality of the cam... - https://hackaday.com/2023/03/28/kino-wheels-gives-you-a-hand-learning-camera-operation/ #arduinomega2560 #mischacks #handwheel #simulator #bearing #encoder #tilt #din #pan
#pan #din #tilt #encoder #bearing #simulator #handwheel #mischacks #arduinomega2560
Zhou et al. propose a noise-aware #deeplearning model 'NAFSA-Net' for #underwater acoustic signal #denoising which adopts #encoder & stacked fullband-subband attention (FSA) blocks to capture both global and fine-grained local dependencies of features
https://ieeexplore.ieee.org/abstract/document/10064313
#deeplearning #underwater #denoising #encoder
Sequential Query Encoding For Complex Query Answering on Knowledge Graphs
Shutter Encoder es un programa que me ha servido para transformar archivos .MOV en .WEBM con alfa. No sé si salvará la vida a alguien, pero a mí me lo ha hecho.
I thought I was ready to fight the final boss of the stream, but x264 had other plans for me.
The MajorLinux Show: Horizon Zero Dawn (Part 3) (01/11/2023)
https://majorlinux.com/the-majorlinux-show-horizon-zero-dawn-part-3-01-11-2023/
#x264 #Video #Encoder #h264 #OBSStudio #NVENC #Nvidia #HorizonZeroDawn #GuerillaGames #Sony #Nixxes #Streaming #Twitch #YouTube #BlackFediverse #BlackFedi #BlackMastodon #Blerd #PC #Gaming
#x264 #video #encoder #h264 #obsstudio #nvenc #nvidia #horizonzerodawn #guerillagames #sony #nixxes #streaming #twitch #youtube #blackfediverse #blackfedi #BlackMastodon #blerd #PC #gaming
Stacking Diverse Architectures to Improve Machine Translation
Andrea Schioppa, Nal Kalchbrenner
I thought I was ready to fight the final boss of the stream, but x264 had other plans for me.
The MajorLinux Show: Horizon Zero Dawn (Part 3) (01/11/2023)
https://majorlinux.com/the-majorlinux-show-horizon-zero-dawn-part-3-01-11-2023/
#x264 #Video #Encoder #h264 #OBSStudio #NVENC #Nvidia #HorizonZeroDawn #GuerillaGames #Sony #Nixxes #Streaming #Twitch #YouTube #BlackFediverse #BlackFedi #BlackMastodon #Blerd #PC #Gaming
#x264 #video #encoder #h264 #obsstudio #nvenc #nvidia #horizonzerodawn #guerillagames #sony #nixxes #streaming #twitch #youtube #blackfediverse #blackfedi #BlackMastodon #blerd #PC #gaming