Papers

Introduction Imagine being a teaching assistant for a university course. Now, imagine 1,028 students just submitted an essay assignment. Even if you spent only five minutes grading each one, that is over 85 hours of non-stop grading. This scalability bottleneck is one of the oldest problems in higher education. With the rise of Large Language Models (LLMs), a solution seems obvious: why not let the AI grade the assignments? We know models like GPT-4 are capable of sophisticated reasoning and analysis. However, moving this technology from a controlled experiment into a real-world classroom is fraught with challenges. How do students feel about being graded by a robot? Who pays for the API costs? And, perhaps most importantly, will students try to trick the AI into giving them a perfect score? ...

Imagine you have just released a new software library or a specialized database API. You want developers to be able to use it effortlessly, perhaps by simply typing natural language commands like “find all users who signed up yesterday” rather than writing complex SQL queries or function calls. To build a tool that translates English into your specific code, you typically face a massive hurdle: data. Training a model to understand a specific API usually requires thousands of pairs of natural language commands and their corresponding code snippets. Creating this dataset is expensive, time-consuming, and tedious. ...

Introduction In recent years, a fascinating intersection has emerged between artificial intelligence and neuroscience. Large Language Models (LMs)—the technology behind systems like GPT—have demonstrated an uncanny ability to predict human brain activity. When a human reads a book inside an fMRI scanner, the internal activations of an LM processing that same text can be mapped surprisingly well to the biological signals in the human’s brain. This phenomenon has sparked a major debate: Why do they align? ...

Introduction Content moderation has come a long way. If you post a slur or a blatantly violent threat on social media today, there is a high probability that an automated system will flag it and remove it within hours. The algorithms trained to spot explicit keywords are efficient. However, hate speech is evolving. It is becoming quieter, subtler, and more insidious. Consider the difference between a direct insult and a sarcastic remark that relies on a shared, negative stereotype. The former is easy for a machine to catch; the latter requires cultural context and reasoning capabilities that most models lack. This is the domain of implicit hate speech. ...

Introduction Large Language Models (LLMs) like GPT-4 and Claude are incredibly proficient at generating human-like text, writing poetry, and even explaining complex historical events. However, there is a specific domain where these models often stumble: algorithmic reasoning. Algorithmic reasoning isn’t just about answering a question; it’s about adhering to a strict set of rules, decomposing a complex problem into a sequence of steps, and maintaining the state of variables throughout the process. A classic example is a logic puzzle involving multiple people lying or telling the truth (a “Web of Lies”), or navigating a grid based on a sequence of turns and steps. ...

How AI Learns the Unseen: The Mystery of the “Beautiful Five Days” Imagine you are reading a book and you come across the phrase: “a beautiful five days.” To a native English speaker, this sounds perfectly natural. You might say, “We spent a beautiful five days in Rome.” But if you pause and look at the grammar, something strange is happening. The word “a” is a singular article (used for one thing, like “a dog”). The phrase “five days” is plural. In strict grammatical terms, combining a singular article with a plural noun phrase should be a disaster. We don’t say “a days” or “a five dogs.” Yet, the “Article + Adjective + Numeral + Noun” (AANN) construction is perfectly acceptable English. ...

Imagine you are looking for instructions on how to bake a specific type of pastry. You type your query into a search engine. Somewhere out there, the perfect recipe exists, written by a master baker. However, that baker wrote the recipe in Italian, and you searched in English. Ideally, a modern semantic search engine should find that document. After all, “flour” means “farina,” and the semantic intent of baking is the same regardless of the language used to describe it. If we have a translation tool, the language of the document shouldn’t matter—only the meaning should. ...

Language barriers are arguably the biggest obstacle to global communication, and for a long time, Machine Translation (MT) has been the battering ram trying to break them down. In recent years, Large Language Models (LLMs) like GPT-4 have revolutionized this field, offering translations that are not just accurate but contextually rich. However, there is a catch. To get top-tier translation performance, you typically have two options: Use a massive, general-purpose LLM (like GPT-4): This yields excellent results but comes with exorbitant infrastructure and deployment costs. Train a translation-specific LLM (like ALMA): This involves pre-training on billions of tokens and fine-tuning on millions of high-quality, human-annotated translation pairs. This is resource-intensive and expensive due to the need for human labor. This creates a significant gap. Is it possible to take a smaller, open-source model and boost its translation capabilities to rival the giants, without spending a fortune on human annotation or massive compute? ...

Imagine trying to explain a complex scientific concept to a 5-year-old, then to a 10-year-old, and finally to a high schooler. You would change your vocabulary, sentence structure, and tone for each “target” audience. This is the essence of Target-level Sentence Simplification. While humans do this naturally, teaching machines to generate text at specific complexity levels (like “Grade 3” vs. “Grade 8”) is notoriously difficult. The primary bottleneck isn’t the model architecture; it’s the data. We simply don’t have enough high-quality parallel datasets—pairs of complex sentences aligned with their simplified versions across multiple grade levels. ...

Large Language Models (LLMs) like GPT-4 and Gemini have transformed how we interact with information. We ask them to write emails, summarize complex topics, and even generate biographies of historical figures. But there is a well-known catch: hallucinations. An LLM can speak with absolute confidence while fabricating facts entirely. For simple “Yes/No” questions or multiple-choice classifications, determining if a model is uncertain is relatively straightforward. We can look at the probability scores (logits) of the output tokens. But how do we measure confidence when the model generates a 300-word biography? If the model writes three paragraphs about a disease, how do we know which sentences are factual and which are creative fiction? ...

Introduction In the rapidly evolving world of Large Language Models (LLMs), the race for “context window” supremacy has been fierce. We’ve gone from models that could barely hold a conversation history to behemoths like GPT-4 and Claude 2, which boast context windows of 128k and 200k tokens respectively. Ideally, this means you should be able to feed a model an entire novel, a legal repository, or a massive technical manual, and ask it any question. ...

The Cheat Sheet Strategy: How LLoCO Masters Long Contexts Efficiently Imagine you are a student preparing for a grueling final exam covering an entire textbook. You have three ways to tackle this. First, the “Open Book” approach: you bring the entire textbook into the exam hall. You have all the information, but flipping through thousands of pages to find one specific answer takes forever. Second, the “Closed Book” approach: you rely solely on what you memorized. It’s fast, but if the exam asks about specific details from page 342, you’re out of luck. ...

Introduction In the current landscape of Artificial Intelligence, scaling laws have ruled supreme: if you want a smarter model, you make it bigger. Models have ballooned from millions to billions, and now trillions of parameters. However, we are hitting a wall. The sheer computational cost of running inference on these massive dense models is becoming unsustainable for many researchers and applications. Enter the Mixture-of-Experts (MoE) architecture. MoE promises the best of both worlds: it decouples the model’s total size (capacity) from its computational cost (inference latency). By only activating a tiny fraction of the network for each token it processes, an MoE model can have the knowledge capacity of a giant model while running with the speed of a much smaller one. ...

When we think of Large Language Models (LLMs) like LLaMA or GPT-4, we usually think of them as masters of language. They write poetry, summarize emails, and debug code. But at their core, these models are sequence predictors—they look at a stream of tokens and predict what comes next. This raises a fascinating question: If the sequence isn’t text, but data from a physical system, can the LLM learn the laws of physics just by looking at the numbers? ...

The world of academic research is facing a crisis of scale. Every year, the number of paper submissions to top-tier Artificial Intelligence conferences skyrockets. For the researchers on the receiving end, this means an ever-growing pile of papers to read, critique, and review. It is a workload that is becoming unsustainable. Enter Large Language Models (LLMs). We know they can write poetry, debug code, and pass the bar exam. Naturally, the question arises: Can LLMs help relieve the burden of peer review? ...

Introduction Imagine you are a simultaneous interpreter at a high-stakes medical conference. A speaker rushes to the podium and begins talking rapidly about cardiology. They mention a patient suffering from “PVC.” If you are just translating word-for-word, you might stumble. Is it Polyvinyl Chloride (a plastic)? No, in this context, it stands for Premature Ventricular Contraction. To make that distinction instantly, you need context. You need to know the topic is cardiology. You might even have a glossary prepared beforehand. ...

Introduction Imagine you are a librarian tasked with organizing a massive pile of books into genres. Most books are easy: the ones with spaceships go into Science Fiction, and the ones with dragons go into Fantasy. But what do you do with a book about a dragon flying a spaceship? Or a book with a torn cover and a vague title? In the world of data science, this is the problem of Text Clustering. It is a fundamental task where we group similar documents together without having any prior labels. While it sounds simple, it is notoriously difficult to get right. Traditional mathematical methods often choke on “edge points”—the ambiguous data points or outliers that don’t fit neatly into a circle. On the other hand, modern Large Language Models (LLMs) like GPT-4 are brilliant at understanding these nuances, but using them to read every single document in a massive dataset is prohibitively expensive and slow. ...

Imagine you find an old executable file on a server. It’s a critical piece of legacy software for your company, but there’s one problem: the source code is gone. No GitHub repo, no backup zip files. Just the raw binary. To update or fix this software, you need to perform decompilation—the process of reversing the compilation steps to turn machine code back into a human-readable programming language like C. For decades, this has been the realm of specialized tools like Ghidra or IDA Pro. While powerful, these tools often produce output that looks more like a mathematical riddle than clean code. ...

Imagine you are an English learner whose native language is Japanese. You are chatting with a friend and you type: “According to the test, my shortcomings are 靴下 and ご主人様.” To a bilingual speaker, this sentence makes perfect sense. It’s a classic example of Code-Switching (CSW)—the fluid alternation between two or more languages in a single conversation. It is a sign of linguistic competence, not confusion. However, if you feed that sentence into a standard Grammatical Error Correction (GEC) system (like the ones powering your favorite writing assistants), it will likely fail. It might flag the Japanese characters as “spelling errors,” try to delete them, or hallucinate English words in their place. ...

Introduction Imagine you are taking a difficult math exam. On the first question, you struggle, make a guess, and get it wrong. But immediately after, you are shown the correct solution. When you encounter a similar problem five questions later, you recall that solution, apply the logic, and get it right. You are learning from experience. Now, consider how we evaluate Large Language Models (LLMs). We typically use benchmarks like MMLU or GSM8K, where models answer thousands of questions. However, in these standard evaluations, every question is treated as an isolated event. The model doesn’t “remember” solving question #1 when it tackles question #100. It doesn’t get to learn from its successes or failures during the test. ...