On Language Generation in the Limit with Bounded Memory

We study language generation in the limit under bounded memory. In this task, a learner observes examples from an unknown target language one at a time and must eventually output only new valid examples. Prior work assumes access to the entire history, a strong assumption since realistic algorithms retain limited past information. Classical work in learning theory shows memory constraints dramatically alter learnability; we extend this to language generation. First, we study memoryless generators. Under a mild enumeration restriction, every countable collection of infinite languages remains generable without memory. Without this restriction, we exactly characterize when memoryless generation is possible. For finite collections, we characterize the optimal minimax density achievable by memoryless generators -- the best density guaranteed against any collection of a given size. This combinatorial bound relies on Sperner's theorem and symmetric chain decompositions. We further show that a sliding window of the last $W$ examples does not improve this worst-case density, whereas allowing it to store $b$ adaptively chosen past examples improves the achievable density for every $b \geq 1$. Finally, we revisit identification in the limit, where the learner must converge to a single correct hypothesis for the target language. We focus on its incremental variant, where the learner remembers only its previous guess. Here, although exact identification fails on a collection of just three languages, a mild relaxation requiring convergence to an ``approximate'' version of the target is achievable for every finite collection. These results show bounded memory affects these tasks differently: generation remains achievable for every countable collection, while density and identification are confined to finite collections, with guarantees weakening as the collection grows.

What is Learnable in Valiant's Theory of the Learnable?

Valiant's 1984 paper is widely credited with introducing the PAC learning model, but it, in fact, introduced a different model: unlike PAC learning, the learner receives only positives, may issue membership queries, and must output a hypothesis with no false positives. Prior work characterized variants, including the case without queries. We revisit Valiant's original model and ask: *Which classes are learnable in it?* For every finite domain, including Valiant's Boolean-hypercube setting, we show that a class is learnable if and only if every realizable positive sample can be certified by a poly-size adaptive query-compression scheme. This is a new variant of sample compression where the learner certifies samples via a short interaction with the membership oracle. Our characterization shows that learnability in Valiant's model is strictly sandwiched between learnability in the PAC model and the variant of Valiant's model without membership queries. This is one of the rare cases where introducing membership queries changes the set of learnable classes, and not just the sample or computational complexity. Next, we study the natural extension of the model to arbitrary domains. While we do not obtain an exact characterization, our techniques readily generalize and show that the same strict sandwiching persists. Finally, we show that $d$-dimensional halfspaces, which are not learnable without queries, are learnable with queries: we give a $\mathrm{poly}(d) \tilde{O}(1/ε)$ sample and $\mathrm{poly}(d) \mathrm{polylog}(1/ε)$ query algorithm, and prove that at least $Ω(d)$ samples or queries are necessary. To our knowledge, this is the first algorithm for halfspaces in Valiant's model. Together, these results uncover a surprisingly rich theory behind Valiant's original notion of learnability and introduce ideas that may be of independent interest in learning theory.

Differentially Private Language Generation and Identification in the Limit

We initiate the study of language generation in the limit, a model recently introduced by Kleinberg and Mullainathan [KM24], under the constraint of differential privacy. We consider the continual release model, where a generator must eventually output a stream of valid strings while protecting the privacy of the entire input sequence. Our first main result is that for countable collections of languages, privacy comes at no qualitative cost: we provide an $\varepsilon$-differentially-private algorithm that generates in the limit from any countable collection. This stands in contrast to many learning settings where privacy renders learnability impossible. However, privacy does impose a quantitative cost: there are finite collections of size $k$ for which uniform private generation requires $Ω(k/\varepsilon)$ samples, whereas just one sample suffices non-privately. We then turn to the harder problem of language identification in the limit. Here, we show that privacy creates fundamental barriers. We prove that no $\varepsilon$-DP algorithm can identify a collection containing two languages with an infinite intersection and a finite set difference, a condition far stronger than the classical non-private characterization of identification. Next, we turn to the stochastic setting where the sample strings are sampled i.i.d. from a distribution (instead of being generated by an adversary). Here, we show that private identification is possible if and only if the collection is identifiable in the adversarial model. Together, our results establish new dimensions along which generation and identification differ and, for identification, a separation between adversarial and stochastic settings induced by privacy constraints.

Language Generation with Infinite Contamination

We study language generation in the limit, where an algorithm observes an adversarial enumeration of strings from an unknown target language $K$ and must eventually generate new, unseen strings from $K$. Kleinberg and Mullainathan [KM24] proved that generation is achievable in surprisingly general settings. But their generator suffers from ``mode collapse,'' producing from an ever-smaller subset of the target. To address this, Kleinberg and Wei [KW25] require the generator's output to be ``dense'' in the target language. They showed that generation with density, surprisingly, remains achievable at the same generality. Both results assume perfect data: no noisy insertions and no omissions. This raises a central question: how much contamination can generation tolerate? Recent works made partial progress on this question by studying (non-dense) generation with either finite amounts of noise (but no omissions) or omissions (but no noise). We characterize robustness under contaminated enumerations: 1. Generation under Contamination: Language generation in the limit is achievable for all countable collections iff the fraction of contaminated examples converges to zero. When this fails, we characterize which collections are generable. 2. Dense Generation under Contamination: Dense generation is strictly less robust to contamination than generation. As a byproduct, we resolve an open question of Raman and Raman [ICML25] by showing that generation is possible with only membership oracle access under finitely many contaminated examples. Finally, we introduce a beyond-worst-case model inspired by curriculum learning and prove that dense generation is achievable even with infinite contamination provided the fraction of contaminated examples converges to zero. This suggests curriculum learning may be crucial for learning from noisy web data.

(Im)possibility of Automated Hallucination Detection in Large Language Models

Is automated hallucination detection possible? In this work, we introduce a theoretical framework to analyze the feasibility of automatically detecting hallucinations produced by large language models (LLMs). Inspired by the classical Gold-Angluin framework for language identification and its recent adaptation to language generation by Kleinberg and Mullainathan, we investigate whether an algorithm, trained on examples drawn from an unknown target language $K$ (selected from a countable collection) and given access to an LLM, can reliably determine whether the LLM's outputs are correct or constitute hallucinations. First, we establish an equivalence between hallucination detection and the classical task of language identification. We prove that any hallucination detection method can be converted into a language identification method, and conversely, algorithms solving language identification can be adapted for hallucination detection. Given the inherent difficulty of language identification, this implies that hallucination detection is fundamentally impossible for most language collections if the detector is trained using only correct examples from the target language. Second, we show that the use of expert-labeled feedback, i.e., training the detector with both positive examples (correct statements) and negative examples (explicitly labeled incorrect statements), dramatically changes this conclusion. Under this enriched training regime, automated hallucination detection becomes possible for all countable language collections. These results highlight the essential role of expert-labeled examples in training hallucination detectors and provide theoretical support for feedback-based methods, such as reinforcement learning with human feedback (RLHF), which have proven critical for reliable LLM deployment.

On Agnostic PAC Learning in the Small Error Regime

Binary classification in the classic PAC model exhibits a curious phenomenon: Empirical Risk Minimization (ERM) learners are suboptimal in the realizable case yet optimal in the agnostic case. Roughly speaking, this owes itself to the fact that non-realizable distributions $\mathcal{D}$ are simply more difficult to learn than realizable distributions -- even when one discounts a learner's error by $\mathrm{err}(h^*_{\mathcal{D}})$, the error of the best hypothesis in $\mathcal{H}$ for $\mathcal{D}$. Thus, optimal agnostic learners are permitted to incur excess error on (easier-to-learn) distributions $\mathcal{D}$ for which $\tau = \mathrm{err}(h^*_{\mathcal{D}})$ is small. Recent work of Hanneke, Larsen, and Zhivotovskiy (FOCS `24) addresses this shortcoming by including $\tau$ itself as a parameter in the agnostic error term. In this more fine-grained model, they demonstrate tightness of the error lower bound $\tau + \Omega \left(\sqrt{\frac{\tau (d + \log(1 / \delta))}{m}} + \frac{d + \log(1 / \delta)}{m} \right)$ in a regime where $\tau > d/m$, and leave open the question of whether there may be a higher lower bound when $\tau \approx d/m$, with $d$ denoting $\mathrm{VC}(\mathcal{H})$. In this work, we resolve this question by exhibiting a learner which achieves error $c \cdot \tau + O \left(\sqrt{\frac{\tau (d + \log(1 / \delta))}{m}} + \frac{d + \log(1 / \delta)}{m} \right)$ for a constant $c \leq 2.1$, thus matching the lower bound when $\tau \approx d/m$. Further, our learner is computationally efficient and is based upon careful aggregations of ERM classifiers, making progress on two other questions of Hanneke, Larsen, and Zhivotovskiy (FOCS `24). We leave open the interesting question of whether our approach can be refined to lower the constant from 2.1 to 1, which would completely settle the complexity of agnostic learning.

Characterizations of Language Generation With Breadth

We study language generation in the limit, introduced by Kleinberg and Mullainathan [KM24], building on classical works of Gold [Gol67] and Angluin [Ang79]. [KM24] proposed an algorithm that generates strings from any countable language collection in the limit. While their algorithm eventually outputs strings from the target language $K$, it sacrifices breadth, i.e., the ability to generate all strings in $K$. A key open question in [KM24] is whether this trade-off between consistency and breadth is inherrent. Recent works proposed different notions of consistent generation with breadth. Kalavasis, Mehrotra, and Velegkas [KVM24] introduced three definitions: generation with exact breadth, approximate breadth, and unambiguous generation. Concurrently and independently, Charikar and Pabbaraju [CP24a] proposed exhaustive generation. Both works examined when generation with these notions of breadth is possible. Building on [CP24a, KVM24], we fully characterize language generation for these notions and their natural combinations. For exact breadth, we provide an unconditional lower bound, removing a technical condition from [KVM24] and extending the result of [CP24a] that holds for specific collections of languages. We show that generation with exact breadth is characterized by Angluin's condition for identification. We further introduce a weaker version of Angluin's condition that tightly characterizes both approximate breadth and exhaustive generation, proving their equivalence. Additionally, we show that unambiguous generation is also characterized by Angluin's condition as a special case of a broader result. Finally, we strengthen [KVM24] by giving unconditional lower bounds for stable generators, showing that Angluin's condition characterizes the previous breadth notions for stable generators. This shows a separation between stable and unstable generation with approximate breadth.

Procurement Auctions via Approximately Optimal Submodular Optimization

We study procurement auctions, where an auctioneer seeks to acquire services from strategic sellers with private costs. The quality of services is measured by a submodular function known to the auctioneer. Our goal is to design computationally efficient procurement auctions that (approximately) maximize the difference between the quality of the acquired services and the total cost of the sellers, while ensuring incentive compatibility (IC), individual rationality (IR) for sellers, and non-negative surplus (NAS) for the auctioneer. Our contributions are twofold: (i) we provide an improved analysis of existing algorithms for non-positive submodular function maximization, and (ii) we design efficient frameworks that transform submodular optimization algorithms into mechanisms that are IC, IR, NAS, and approximation-preserving. These frameworks apply to both the offline setting, where all sellers' bids and services are available simultaneously, and the online setting, where sellers arrive in an adversarial order, requiring the auctioneer to make irrevocable decisions. We also explore whether state-of-the-art submodular optimization algorithms can be converted into descending auctions in adversarial settings, where the schedule of descending prices is determined by an adversary. We show that a submodular optimization algorithm satisfying bi-criteria $(1/2, 1)$-approximation in welfare can be effectively adapted to a descending auction. Additionally, we establish a connection between descending auctions and online submodular optimization. Finally, we demonstrate the practical applications of our frameworks by instantiating them with state-of-the-art submodular optimization algorithms and empirically comparing their welfare performance on publicly available datasets with thousands of sellers.

Randomized Truthful Auctions with Learning Agents

We study a setting where agents use no-regret learning algorithms to participate in repeated auctions. \citet{kolumbus2022auctions} showed, rather surprisingly, that when bidders participate in second-price auctions using no-regret bidding algorithms, no matter how large the number of interactions $T$ is, the runner-up bidder may not converge to bidding truthfully. Our first result shows that this holds for \emph{general deterministic} truthful auctions. We also show that the ratio of the learning rates of the bidders can \emph{qualitatively} affect the convergence of the bidders. Next, we consider the problem of revenue maximization in this environment. In the setting with fully rational bidders, \citet{myerson1981optimal} showed that revenue can be maximized by using a second-price auction with reserves.We show that, in stark contrast, in our setting with learning bidders, \emph{randomized} auctions can have strictly better revenue guarantees than second-price auctions with reserves, when $T$ is large enough. Finally, we study revenue maximization in the non-asymptotic regime. We define a notion of {\em auctioneer regret} comparing the revenue generated to the revenue of a second price auction with truthful bids. When the auctioneer has to use the same auction throughout the interaction, we show an (almost) tight regret bound of $\smash{\widetilde \Theta(T^{3/4})}.$ If the auctioneer can change auctions during the interaction, but in a way that is oblivious to the bids, we show an (almost) tight bound of $\smash{\widetilde \Theta(\sqrt{T})}.$

On the Limits of Language Generation: Trade-Offs Between Hallucination and Mode Collapse

Specifying all desirable properties of a language model is challenging, but certain requirements seem essential. Given samples from an unknown language, the trained model should produce valid strings not seen in training and be expressive enough to capture the language's full richness. Otherwise, outputting invalid strings constitutes "hallucination," and failing to capture the full range leads to "mode collapse." We ask if a language model can meet both requirements. We investigate this within a statistical language generation setting building on Gold and Angluin. Here, the model receives random samples from a distribution over an unknown language K, which belongs to a possibly infinite collection of languages. The goal is to generate unseen strings from K. We say the model generates from K with consistency and breadth if, as training size increases, its output converges to all unseen strings in K. Kleinberg and Mullainathan [KM24] asked if consistency and breadth in language generation are possible. We answer this negatively: for a large class of language models, including next-token prediction models, this is impossible for most collections of candidate languages. This contrasts with [KM24]'s result, showing consistent generation without breadth is possible for any countable collection of languages. Our finding highlights that generation with breadth fundamentally differs from generation without breadth. As a byproduct, we establish near-tight bounds on the number of samples needed for generation with or without breadth. Finally, our results offer hope: consistent generation with breadth is achievable for any countable collection of languages when negative examples (strings outside K) are available alongside positive ones. This suggests that post-training feedback, which encodes negative examples, can be crucial in reducing hallucinations while limiting mode collapse.