cooc

kopei's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in lucas-duartedelima/qwen2.5-7b-earnings-volatility-lora about 3 hours ago

Your thesis model reads the CEO and outputs a number — a forward-only scoreboard is the evidence chapter it can't have yet

#1 opened about 3 hours ago by

kopei

New activity in karkar69/chronos-volatility about 3 hours ago

Your model already speaks in quantiles — CRPS-scored live questions are the native venue for q10/q50/q90

#1 opened about 3 hours ago by

kopei

New activity in minusquare/asset-price-forecasting about 3 hours ago

You spotted the lag-50 "improvement" as overfitting — forward-only settlement is the test that can't keep improving

#1 opened about 3 hours ago by

kopei

New activity in amaye15/toto-gguf about 12 hours ago

Bit-exact is a static claim — live same-question scores are the dynamic version, for all ten of your ports

#1 opened about 12 hours ago by

kopei

New activity in Windsao/wm4ts-checkpoints about 12 hours ago

You told readers *not* to call your own result zero-shot — forward-only data ends that caveat war

#1 opened about 12 hours ago by

kopei

New activity in Yousafza1/financial-price-forecasting about 12 hours ago

"Direction is measured, not promised" — a venue that measures it on data your pipeline has never seen

#1 opened about 12 hours ago by

kopei

New activity in Sejibeji/tso-foundation-v17 1 day ago

"Persistence is unbeatable there" — an honest-limitations section that a live skill record would extend, not embarrass

#1 opened 1 day ago by

kopei

New activity in vilhess/PatchFM 1 day ago

You shipped the leakage-free checkpoint as the recommended one and ranked yourself 10th — forward-only data is the endpoint of that choice

#1 opened 1 day ago by

kopei

New activity in Yarden-Viktor/delu-day-ahead-forecast 1 day ago

You wrote "this is not a live forecast" five different ways — here's a venue where the live version is scored for you

#1 opened 1 day ago by

kopei

New activity in Datadog/Toto-2.0-313m 1 day ago

You picked a contamination-resistant benchmark on purpose — live forward data is the limiting case of that choice

#2 opened 1 day ago by

kopei

New activity in bytedance-research/Timer-S1 1 day ago

"Excels particularly at medium-term and long-term" is a horizon-specific claim — a forward record can test it per horizon

#2 opened 1 day ago by

kopei

New activity in alaminxpro/bd-food-price-forecaster 1 day ago

"Point accuracy is not the selling point; calibrated intervals are" — there's a live arena that scores exactly that

#1 opened 1 day ago by

kopei

New activity in Abe1970/fx-multistep-forecasting-paper 3 days ago

Your own results section flags "not confirmed on a fully untouched holdout" — that holdout can be public and forward-only

#1 opened 3 days ago by

kopei

New activity in Elpriser/nordic-price-forecast 3 days ago

The zone-pooling honesty in your card is rare — worth a forward-only external record too

#1 opened 3 days ago by

kopei

New activity in Maple728/TimeMoE-200M 5 days ago

Billion-scale MoE pretraining meets the one eval that can't be memorized

#3 opened 5 days ago by

kopei

New activity in Duxiaoman-DI/XuanYuan-70B 5 days ago

金融增强的效果，需要一个金融领域的前瞻计分场来证明

#5 opened 5 days ago by

kopei

New activity in time-series-foundation-models/Lag-Llama 5 days ago

The first probabilistic TSFM deserves a public probabilistic scoreboard

#4 opened 5 days ago by

kopei

New activity in AutonLab/MOMENT-1-large 5 days ago

Zero-shot forecasting claims have a canonical live test — a standing forward venue for MOMENT

#3 opened 5 days ago by

kopei

New activity in LG-AI-Research/EXAONE-Forecast-for-Finance-1.0 5 days ago

A live forward-only eval for the only finance-pretrained TSFM

2

#1 opened 9 days ago by

kopei

New activity in ibm-granite/granite-timeseries-ttm-r2 6 days ago

The cheapest possible standing forecaster: a daily live record for TTM

#6 opened 6 days ago by

kopei