Cover image

A Different Forecast Without a Different Model

TL;DR for operators A utility forecasting hundreds of heterogeneous customers and distribution transformers faces a model-management decision: maintain one shared forecaster, train separate local models, or specialize a common model without multiplying its full parameter set. Li, Cheng, and Weng demonstrate a third architecture. Their method keeps a probabilistic forecasting backbone fixed and constructs customer-specific predictive mean and variance heads from a shared bank of low-rank adaptations. Customers can combine several adapters according to their load characteristics rather than receiving an independently trained model. ...

November 11, 2026 · 8 min · Zelina
Cover image

The Detector Can’t See What Its Features Don’t Describe

TL;DR for operators When an anomaly detector misses events that matter, teams often look first at the model, the feature selector, or the score-combination rule. The results in LLM-Generated Feature Pools for Time Series Anomaly Detection1 point to an earlier constraint: what the detector is allowed to measure. In matched experiments on the univariate TSB-AD-U benchmark, changing the candidate feature pool spans 0.226 VUS-PR, compared with 0.031 across feature-selection strategies even after adding a hindsight oracle, and 0.096 across the full score-aggregation grid. A 10-feature hand-crafted pool reaches 0.529, while a stronger LLM-generated pool reaches 0.569 on average. Combining generated and hand-crafted features raises the corresponding result to 0.588. ...

October 8, 2026 · 7 min · Zelina
Cover image

When the Test Window Changes the Problem

TL;DR for operators A forecasting benchmark can change the problem being measured without changing the nominal dataset. In the rideshare data examined here, zeros make up 46.9% of the full dataset but only 5.3% of the standard rolling-origin evaluation windows. Under a series-wise split, the evaluation-window zero rate rises to 59.1%, and the interpretation of an autoregressive hurdle model reverses. ...

August 20, 2026 · 7 min · Zelina
Cover image

Important, but Not Direct: When Time-Series Attribution Misstates Model Dependencies

TL;DR for operators A forecasting dashboard can correctly report that an earlier observation influenced a prediction and still give the wrong impression about how that influence enters the model. Amadeo Tunyi’s paper, The Failures of Marginal Influence-Based Attribution Methods for Global Time Series Explanations, argues that familiar scalar attribution methods cannot in general recover the model’s direct temporal dependency structure.1 Marginal methods can assign importance to an earlier variable whose influence is entirely mediated through a later, autocorrelated observation. Gradient methods can report sensitivity that exists only outside the support of the data the model actually sees. ...

August 19, 2026 · 8 min · Zelina
Cover image

Seeing Is Judging: Why LLMs Are Better Critics Than Creators in Time-Series Reasoning

A dashboard says revenue demand has “stabilized.” A monitoring agent says a sensor spike is “temporary.” A trading assistant says volatility has “fallen after the regime shift.” The sentence is smooth. The chart is nearby. The user is tired. That is usually enough for a bad explanation to survive. This is the quiet problem behind AI-assisted analytics: not whether a language model can write a plausible story about time-series data, but whether the story is faithful to the numbers. A recent paper, LLM-as-a-Judge for Time Series Explanations, studies exactly this gap by asking models to play two different roles: narrator and critic.1 ...

April 4, 2026 · 16 min · Zelina
Cover image

When LLMs Meet Time: Why Time-Series Reasoning Is Still Hard

Dashboard numbers are seductive because they look obedient. Revenue goes up, traffic dips, latency spikes, inventory turns over, temperature drifts, volatility clusters. Put the sequence into a chart and the pattern seems almost polite. Then someone asks an LLM what happened. The model answers fluently. It may even sound like an analyst who has seen too many quarterly review decks and has developed a protective layer of confidence. But fluency is not temporal understanding. A model can describe a curve, name a trend, and still fail to understand which segment comes next, whether a transformation is correct, or whether a discontinuity is an error or a legitimate feature of the process. ...

February 3, 2026 · 16 min · Zelina
Cover image

Rationales Before Results: Teaching Multimodal LLMs to Actually Reason About Time Series

Dashboard work has a familiar little ritual. Someone opens a chart, zooms into the last few points, notices a dip, a rebound, or a suspiciously clean trend line, and then says something that sounds analytical: “Looks like it will continue.” Sometimes that is wisdom. Sometimes it is just a human staring confidently at a squiggle. ...

January 7, 2026 · 15 min · Zelina
Cover image

Kill the Correlation, Save the Grid: Why Energy Forecasting Needs Causality

Humidity looks harmless on a scatter plot. Actually, in this paper, it looks worse than harmless: it appears negatively correlated with electricity demand. That is the kind of result a busy forecasting team might quietly accept. Add humidity as a feature, let the model figure it out, move on. The grid will not wait politely while everyone debates Pearl diagrams. ...

December 15, 2025 · 14 min · Zelina
Cover image

HAROOD: When Benchmarks Grow Up and Models Stop Cheating

A wearable model can look brilliant in the lab and embarrass itself on Monday morning. The user changes. The watch slides down the wrist. A sensor is mounted on the chest instead of the pocket. The same person walks differently after fatigue, injury, aging, or simply because life has the terrible habit of not matching the training set. Human Activity Recognition, or HAR, has always lived with this problem. It turns sensor streams from accelerometers, gyroscopes, EMG, ECG, and other wearable or ambient devices into labels such as walking, running, sitting, cycling, or stress state. It is useful precisely because it moves into the real world. That is also where benchmark accuracy goes to die. ...

December 12, 2025 · 20 min · Zelina
Cover image

How to Make Neural Networks Talk: Register Automata as Their Unexpected Interpreters

How to Make Neural Networks Talk: Register Automata as Their Unexpected Interpreters Prices move. Sensors drift. Users click, pause, return, disappear, and sometimes behave exactly like a Markov chain with a caffeine problem. Modern sequence models are good at turning such streams into decisions. A recurrent network or transformer can look at a run of numbers and say: buy, flag, reject, approve, alert. What it usually cannot do is explain the rule it has learned in a form that a risk team, engineer, or auditor can actually inspect. ...

November 25, 2025 · 18 min · Zelina